Forms of Deference in Prison Law

It has been almost five decades since the U.S. Supreme
Court first opened the courthouse door to prisoner suits,
putting paid to the then-prevalent notion that the federal
courts had no jurisdiction to hear cases brought by prisoners.2 But even during the short-lived “reform” era that
followed, the Court took pains to emphasize the need for
judicial restraint in this area.3 And in the years since, this
imperative of restraint—aka deference—has emerged as
the strongest theme of the Court’s prisoners’ rights jurisprudence.4
It is not unreasonable for courts to grant a measure of
deference to state actors tasked with a job as complex,
challenging, and hazardous as running the prisons. It
does, after all, seem probable that in many cases, what
may appear inappropriate to a court far removed from the
hothouse of a carceral environment was in fact a reasonable undertaking under the circumstances. It is, however,
also the role of the courts to enforce the Constitution, and
if that role is not to be abdicated, the need for deference
ought not to preclude good faith implementation of constitutional protections.
Of these two imperatives—judicial deference and
meaningful constitutional enforcement—the former is
arguably the primary driver of the Court’s prisoners’
rights jurisprudence. Yet taken as a body, the cases in this
area reveal no principled basis for determining when deference is justified, what forms it may legitimately take, or
the proper limits on its use. Instead, the mere mention of
“deference” has emerged as a catch-all justification for
curtailing both the burden on prison officials to ensure
constitutional prisons and prisoners’ prospects for recovery even for arguably meritorious claims.
Deference to prison officials serves at least three main
functions in the Supreme Court’s prison law cases: it
(1) informs the construction of substantive constitutional
doctrine, (2) justifies the altering of procedural rules, and
(3) frames the interpretation and assessment of relevant
facts. In cases involving prisoners’ constitutional rights,
the Court makes routine and unhesitating use of deference
in all these forms to tilt the balance against prisoners. But
somewhat unexpectedly, in cases involving the Prison Litigation Reform Act of 19955 (PLRA), the signs are more
mixed; in particular, one can find instances in which the

Court rejects at least the latter two strategies. Despite the
plainly deferential cast of the PLRA’s exhaustion requirement, in Jones v. Bock6 the Court rejected efforts by the
Sixth Circuit to modify the procedural rules governing
exhaustion so as to further disadvantage prisoners. And
more recently, in Brown v. Plata,7 the Court refused an invitation by Justice Alito in dissent to reframe the procedural
history of the case in ways sympathetic to defendants8 yet
wholly unsupported by the record. As will be seen, both
these moves—the Sixth Circuit’s crafting of new procedural rules and Justice Alito’s creative reading of the
record—have analogues in the Court’s prisoner’s rights
jurisprudence, where a general “principle of deference”9
has prompted the court to recast procedural rules in defendant-friendly terms and to downplay or ignore altogether
evidence against the state’s position when interpreting the
record. Although in neither Jones nor Plata does the Court
frame its position in terms of the need for limits on deference, the prison law cases as a body strongly suggest that
such limits are necessary if the law in this area is to overcome the impression presently created by the Court’s
prisoners’ rights jurisprudence, that of a skewed process
that deprives a whole category of vulnerable citizens of
meaningful constitutional protections while only seeming
to take their legal claims seriously.
The role of deference in prison law merits a longer and
more in-depth treatment than I am able to offer here. This
essay is simply a first cut at the issue. As such, it does three
things. First, it identifies the three main forms that deference takes in this area and provides examples of the
deployment of each in the prisoners’ rights cases. Second,
it highlights those moments in the PLRA cases where the
Court has opted not to defer (thus seeming to reject strategies it freely employs elsewhere), and offers a preliminary
hypothesis as to why this might be. Third and finally, it
calls for a theory of deference for the prison law context,
i.e., for the development of principles to guide judicial deference in prison law cases and to set appropriate limits on
its use.
A brief word on terminology: This essay refers variously
to the Court’s prisoners’ rights (or prisoners’ constitutional
rights) cases and to its prison law cases. The former is
intended as a subset of the latter, which encompasses all

Sharon
Dolovich 1
Professor of Law,
UCLA School
of Law

Federal Sentencing Reporter, Vol. 24, No. 4, pp. 245–259, ISSN 1053-9867 electronic ISSN 1533-8363.
©2012 Vera Institute of Justice. All rights reserved. Please direct requests for permission to photocopy
or reproduce article content through the University of California Press’s Rights and Permissions website,
http://www.ucpressjournals.com/reprintInfo.asp. DOI: 10.1525/fsr.2012.24.4.245.

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 245

Electronic copy available at: http://ssrn.com/abstract=2043225

245

17/04/12 12:19 PM

cases involving claims by prisoners. The PLRA cases discussed in Part II constitute another subset of the larger
body of case law referred to here as “prison law.” And
because the cases in question are civil actions brought by
prisoner-plaintiffs, the defendants in each instance are
prison officials being sued for violations of prisoners’
constitutional rights.
I.  Deference in Action

In cases concerning prisoners’ constitutional rights, the
imperative of judicial deference takes three main forms.
First and most obviously, deference is doctrine-constructing.
In this aspect, deference to prison officials is written right
into the substantive constitutional standards, yielding
rules of decision that tip the scales in favor of defendants.
Second, deference is used to justify procedural rule-revising
in ways that transform familiar aspects of the legal process
into more inherently defendant-friendly procedural mechanisms. Third and finally, deference spurs situationreframing, the recasting of a procedural or factual history
in ways that enhance the defendant’s position at the
plaintiff’s expense.
There are many examples of deference operating in
each of these ways in the Supreme Court’s prisoners’
rights cases. The most obvious example of deference in its
first, doctrine-constructing form is Turner v. Safley.10 In
Turner, the Court held that prison regulations that burden
prisoners’ constitutional rights may nonetheless be
upheld if they are “reasonably related to legitimate penological interests.”11 This holding is itself deferential,
creating a space in which prison officials can violate constitutional rights if they can show that doing so facilitates
the running of the prison. But the extent to which Turner
writes deference into law is clearest in the Court’s description of the four factors courts are to use to determine
when challenged regulations satisfy the basic Turner standard. The four factors themselves do not necessarily betray
a pro-deference slant: courts must ask (1) whether there is
a “‘valid, rational connection’ between the prison regulation and the legitimate governmental interest put forward
to justify it”;12 (2) whether there are “alternative means of
exercising the right that remain open to prison inmates”;13
(3) what “impact accommodation of the asserted constitutional right will have on guards and other inmates, and on
the allocation of prison resources generally”;14 and (4)
whether there are “ready alternatives” by which prison
officials can realize their interests while also affording
prisoners the exercise of their rights.15
But the Turner Court’s elaboration of each of these factors leaves no doubt that the test is intended to be
extremely deferential, and provides language for lower
courts to draw on to frame this deference as a legal mandate. Thus, the Court explained (1) that “a regulation
cannot be sustained where the logical connection between
the regulation and the asserted goal is so remote as to render the policy arbitrary or irrational”;16 (2) that “where
‘other avenues’ remain available for the exercise of the

246

FSR2404_04.indd 246

right, courts should be particularly conscious of the measure of judicial deference owed to correctional officers”;17
(3) that “in the necessarily closed environment of the correctional institution, few changes will have no ramifications
on the liberty of others or on the use of the prison’s limited resources”;18 and (4) that “if an inmate claimant can
point to an alternative that fully accommodates the prisoner’s rights at de minimis cost to valid penological interests,
a court may consider that as evidence that the regulation
does not satisfy the reasonable relationship standard.”19 In
short, having plainly instructed lower courts that they
must be deferential in assessing alternatives (factor 2) and
that any change to a prison regime will necessarily have
ramifications for the institution (factor 3), the Turner
Court made clear that, unless the challenged policy is
found to be an “arbitrary or irrational” method for the
state to achieve its stated goals (factor 1) and claimants can
identify an alternative means to “fully accommodate” their
rights without any appreciable cost to the prison (factor 4),
the challenged regulation is to be upheld. And sure
enough, it is a rare case decided under Turner in which the
plaintiff ultimately prevails.20
Turner, moreover, is hardly the only case that writes
deference right into the standards. In Whitley v. Albers,21
the Court addressed the question of when the use of force
by prison officials against prisoners violates the Eighth
Amendment. Although previous cases had held that
prison officials violate the Eighth Amendment when they
are “deliberate[ly] indifferen[t]” to prisoners’ basic needs,22
Whitley held that use of force against prisoners is only
unconstitutional when applied “maliciously and sadistically for the very purpose of causing harm.”23 Absent a
showing of this extremely high (and extremely deferential)
mens rea standard, no use of force—however excessive,
however unreasonable—will be found unconstitutional.
And in Farmer v. Brennan,24 the Court defined deliberate
indifference as the equivalent of criminal recklessness,
thereby protecting prison officials from liability even for
conditions, however egregious, about which they should
have known but did not.25 Farmer framed its holding as
necessitated by the language of the Eighth Amendment; in
Wilson v. Seiter,26 the Court had found that “[i]f the pain
inflicted is not formally meted out as punishment by the
statute or the sentencing judge, some [subjective] mental
element must be attributed to the inflicting officer before
it can qualify [as punishment].”27 But Wilson’s reasoning
does not stand up to scrutiny,28 and once this (pretextual)
justification is removed, it is hard not to see Farmer’s holding as a product of the Court’s determination to create a
zone for the exercise of official discretion undisturbed by
constitutional imperatives.
Deference also operates in the Court’s prisoners’ rights
cases in the second way noted above, i.e., to justify the
altering of existing procedural rules in ways benefiting
defendants at plaintiffs’ expense. One early example of
this form of deference in practice is found in the 1977 case
of Jones v. North Carolina Prisoners’ Labor Union, Inc.29

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

Electronic copy available at: http://ssrn.com/abstract=2043225

17/04/12 12:19 PM

At the time, North Carolina permitted prisoners to join a
labor union, but imposed rules prohibiting members from
holding meetings or from soliciting new members either
in person or through bulk mailings from outside the
prison. Union members who broke these rules were subject to discipline. Prisoners belonging to the union
brought suit, alleging violation of their First Amendment
rights of speech and association, and also violation of
Equal Protection on the basis that other prisoner organizations were not subject to the same restrictions.30
Defendants argued that the restrictions were justified on
grounds of prison security—specifically, the concern that
some union spokespeople could “establish a power bloc”
within the prison that “could be utilized to cause work
slowdowns or stoppages or other undesirable concerted
activity.”31 A three-judge panel of the District Court,
although crediting the sincerity of defendants’ beliefs, also
heard expert evidence to the contrary and thus concluded
that the “expert opinion evidence” was “conflicting.”32 Noting that “not one scintilla of evidence” had been offered to
show that the union had ever interfered with the prison’s
operations,33 the panel ultimately found itself unpersuaded by the notion that it was “necessary or essential to
security and order in the prisons to forbid solicitation of
membership in a union permitted by the authorities.”34
Having thus weighed the testimony in its capacity as
finder of fact, the panel found for the prisoners on their
First Amendment and Equal Protection claims.35
The Supreme Court, however, reversed.36 Emphasizing
“the wide-ranging deference to be accorded the decisions
of prison administrators,”37 Justice Rehnquist took issue,
not with the substance of the panel’s reasoning, but with
its approach to weighing the evidence. As Justice
Rehnquist put it:
Without a showing that [defendants’] beliefs were
unreasonable, it was error for the District Court to
conclude that [they] needed to show more. In particular, the burden was not on [defendants] to show
affirmatively that the Union would be detrimental to
proper penological objectives or would constitute a
present danger to security and order. Rather, such
considerations are peculiarly within the province and
professional expertise of corrections officials, and, in
the absence of substantial evidence in the record to indicate that the officials have exaggerated their response to
these considerations, courts should ordinarily defer
to their expert judgment in such matters.38
In other words, although it is ordinarily the job of the
trier of fact to hear witness testimony—including expert
testimony—and to weigh the evidence presented, Justice
Rehnquist made clear that in prison cases, unless there is
“substantial evidence” to suggest that officials’ response
was “exaggerated,” courts are to presume the correctness of
defendants’ assertions.39 This mandated deference is more
than an acknowledgment that people with expertise in running the prisons are likely to have a deeper understanding

of the matter. At trial, the District Court heard direct testimony supporting the prisoners’ position from a range of
experienced prison administrators, including a Rhode
Island prison warden and the Executive Director of the
North Carolina Inmate Grievance Commission (and member of the state Sentencing Commission). It also
considered documentary evidence from a former Minnesota Corrections director who at the time of trial was
Director of Adult Corrections for the state of Delaware.40
In other words, experts on both sides had plenty of experience and expertise. But North Carolina Prisoners’ Union
made it clear that it is the defendants to whom the court
should defer, crediting their testimony absent an affirmative “showing that [their] beliefs were unreasonable.” It
thereby tilted the scales of judicial deliberation strongly
toward prison officials at the outset, signaling that defendant prison officials in § 1983 actions brought by prisoners
are to be freed from the usual expectation that their own
evidence characterizing the situation will be subject to challenge and to meaningful judicial scrutiny.
A more recent example of the altering of established
procedural rules for the prisoners’ rights context in the
name of deference is found in the 2006 case of Beard v.
Banks.41 In Banks, prisoners held in Pennsylvania’s “longterm segregation unit” (LTSU) challenged on First
Amendment grounds a policy denying them access to
newspapers, magazines, and personal photographs. The
District Court, applying the Turner standard, granted summary judgment, and although the Third Circuit reversed,
the Supreme Court sided with the District Court.42 In particular, a plurality of the Court found “a valid, rational
connection” between the challenged policy and the “governmental interest put forward to justify it.”43 According
to defendants, the restriction was necessary to “motivate
better behavior on the part of particularly difficult prisoners” in the LTSU.44 In finding for defendants, the plurality
relied on the testimony of the prison’s deputy superintendent, who said in his deposition that incentives for good
behavior in the control unit were highly limited and that
“the Policy serves to encourage . . . progress and discourage backsliding.”45 It appears from the case that Banks had
poor representation,46 but he nonetheless contested the
motion for summary judgment on the grounds that “contact with the world generally favors rehabilitation.”47 For
its part, the Third Circuit found insufficient evidence that
“the ban was implemented in a way that could modify
behavior” or to suggest that the defendants’ “deprivation
theory of behavior modification had any basis in real
human psychology, or had proven effective with LTSU
inmates.”48 And in dissent, Justice Stevens closely scrutinized the government’s justification in light of the
realities in the unit and made a compelling case on the
facts that summary judgment was inappropriate.49
Because the case involved a grant of summary judgment, the arguments supporting the prisoners’ position
offered by the Third Circuit and in Justice Stevens’s dissent have particular significance. Ordinarily, a grant of

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 247

Electronic copy available at: http://ssrn.com/abstract=2043225

247

17/04/12 12:19 PM

summary judgment is appropriate only when, taking the
evidence in the light most favorable to the non-moving
party, the defendants can show that no triable issue of
material fact exists.50 In cases governed by Turner, summary judgment should therefore be denied if there is
sufficient evidence on which a jury could find that no
“valid, rational connection” existed between the challenged policy and the proffered justifications for it.
Arguably, the Court of Appeals offered reason enough to
deny the motion: given the lack of supporting evidence for
the behavior modification theory, a reasonable jury could
well fail to find such a connection in this case. Yet Justice
Breyer concluded otherwise. Writing for a plurality of the
Court, he found that the Third Circuit placed too high a
burden on the defendants, according “too little deference”
to the judgment of prison officials about such matters.51
This notion may at first seem perplexing: does not the
summary judgment standard require courts to “draw all
justifiable inferences in favor of the nonmoving party”?52
Maybe in the ordinary case. But when the case involves
prisoners, Justice Breyer held, “we must distinguish
between evidence of disputed facts and disputed matters of
professional judgment.”53 And “[i]n response to the latter,
our inferences must accord deference to the views of
prison authorities. Unless a prisoner can point to sufficient evidence regarding such issues of judgment to allow
him to prevail on the merits, he cannot prevail at the summary judgment stage.”54 In other words, in cases brought
by prisoners, before any discovery has even been allowed,
plaintiffs must overcome the strong presumption that
already exists in favor of the defendants’ view of what is
necessary to run the prisons, and convince the court not
just that a reasonable jury could find in plaintiffs’ favor but
that it will necessarily do so.
The Banks holding turns the summary judgment standard on its head, clearing the way for defendants to
prevail, even in the face of strong opposing arguments, so
long as they can provide a colorable argument for their
position. As Justice Ginsburg observed in her Banks dissent, defendants’ showing in this case—based purely on
defendants’ own conclusory statements—“is slim, the
kind that could be made to justify virtually any prison regulation that does not involve physical abuse.”55 Yet the
point of the summary judgment standard, she notes, is to
determine whether defendants are “entitled to a judgment
as a matter of law.”56 This conclusion ought only to be
reached without trial in cases where no reasonable jury
could find otherwise. Yet on the standard as reframed by
Justice Breyer, “[i]t suffices for [prison officials] to say, in
our professional judgment the restriction is warranted.”57
Third and finally, the imperative of judicial deference
operates in prisoners’ rights cases to recast relevant facts
in ways that deny or disregard the lived experiences of
prisoners, thereby undermining the force of the constitutional claim at issue. Some examples: In Rhodes v.
Chapman,58 the Court rejected a per se challenge to the
practice of double-celling (that is, housing two men in

248

FSR2404_04.indd 248

sixty-three-square-foot cells originally intended for one
person) as not in itself the “wanton and unnecessary
infliction of pain” prohibited by the Eighth Amendment.59 The Court maintained that no ground existed for
an Eighth Amendment claim, since the double-celling did
not “create other conditions intolerable for prison confinement.”60 The Court reached this conclusion although
the great weight of the evidence indicated that “a longterm inmate must have to himself” a minimum of fifty
square feet of floor space “in order to avoid serious mental, emotional, and physical deterioration”—and that,
after accounting for the bed alone, even “without making
allowance for any other furniture in the room” (e.g., toilet, sink, locker, shelves. etc.), the remaining square
footage per person was approximately “20–24 square feet,
an area about the size of a typical door.”61 In Whitley v.
Albers, the Court upheld a directed verdict for defendants
on a claim of unconstitutional use of force brought by a
prisoner shot by an officer in the course of a prison disturbance.62 Writing for the Court, Justice O’Connor
emphasized that at the time of the shooting, a riot was in
progress, “a guard was still held hostage,” and “[t]he situation remained dangerous and volatile.”63 Yet as Justice
Marshall noted in dissent, in a case appealing a directed
verdict for the defendants, the “facts must be viewed in
the light most favorable” to the plaintiff, and in this case,
the plaintiff had presented “substantial testimony” at trial
to show that by the time the officer shot him, “the disturbance had subsided,” that “only one prisoner”—a man
named Klenk—“remained in any way disruptive,” and
that the guard who had been taken hostage in the melee
“was not, in fact, in danger,” but “had been put into a cell
by several inmates to prevent Klenk from harming
him.”64 In Overton v. Bazzetta,65 the Court upheld on
Turner grounds a two-year ban on visitors for Michigan
inmates who had two or more substance abuse infractions. Pursuant to Turner, the Court conceded that, “[w]ere
it shown that no alternative means of communication
existed, . . . it would be some evidence that the regulations
were unreasonable,” but then went on to find
“communicat[ion] with persons outside the prison by letter and telephone” to be sufficient alternatives.66 It thus
waved away the seemingly valid concerns that such alternatives are “inadequate for illiterate inmates and for
communications with young children” and that “phone
calls are brief and expensive,” on the telling formalistic
grounds that “[a]lternatives to visits need not be ideal,”
but “need only be available.”67 And perhaps most striking
of all, in Lewis v. Casey,68 Justice Scalia maintained that
prisoners’ right of access to the courts—the right, after
Bounds v. Smith,69 to “adequate law libraries or adequate
assistance from persons trained in the law”70—entails
only the right to “bring to court a grievance that the
inmate wished to present.”71 Justice Scalia’s opinion suggested that anything more than access to this narrow
extent would be well beyond what the Constitution
requires, and he went out of his way to disclaim any

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

language in Bounds that might be read to suggest that
“the State must enable the prisoner to discover grievances,
and to litigate effectively once in court.”72 Yet as anyone
familiar with the legal process is well aware, and as the
Bounds Court emphasized, “[i]t would verge on incompetence for a lawyer to file an initial pleading without
researching such issues as jurisdiction, venue, standing,
exhaustion of remedies, proper parties, [etc.].”73 Moreover, the Bounds Court noted that the state’s response to a
pro se pleading “will undoubtedly contain seemingly
authoritative citations. Without a library, an inmate will
be unable to rebut the State’s argument.”74 Without legal
assistance through the life cycle of a claim, in other
words, it is hard to see how a prisoner’s right of access to
the courts is to have any meaning.
Read separately, there may well be an explanation for
the Court’s position in each of these cases. A per se ban
on double-celling may have been impossible for the
states to implement, making such a holding an expenditure of the Court’s political capital to no purpose.
Finding for the plaintiff in Albers might perhaps have
chilled the ability of prison officials to respond adequately to disturbances in their facilities. Striking down
the ban on visitation for Michigan prisoners with multiple substance-abuse violations might have deprived the
prison of an available incentive for prisoners to avoid
drugs and alcohol while in custody.75 And making the
right of access meaningful over the life cycle of a claim
would surely tax prison systems. Yet in none of these
cases is the Court forthright about these possible implications, which might have allowed explicit consideration
of the legitimacy of allowing such concerns to dictate
constitutional outcomes. Instead, in its quest to reach the
desired result, the Court simply pretends that the facts as
it frames them require the stipulated outcome, reasoning
in ways that not only favor defendants but also seem willfully to deny the lived experience of prisoners—even
when the nature of that experience is the gravamen of
the legal complaint.
In sum, far from achieving a balance between appropriate deference and appropriate constitutional
enforcement, the Court’s prisoners’ rights case law
seems instead to be a jurisprudence of evasion, justified
by talismanic reference to the need to defer. The overall
impression is of an area of constitutional law that preserves the form in each case yet lacks what is arguably
the heart of the enterprise: a good faith assessment of
each claim in light of applicable constitutional principles,
fairly and consistently applied. In an area as fraught as
the running of prisons, it may be reasonable at times to
create space for the exercise of official discretion. But far
from seeking principles for the appropriate scope for
discretion, and thus for an appropriately calibrated
deployment of judicial deference, the Court seems
instead to have recast deference as an all-purpose justification for narrowing the scope for prisoners’ recovery. At
times, the Court’s reasoning in these cases verges on

casuistry—which should trouble anyone committed to
the rule of law.
II. Limiting the Creative Use of Deference:
The Court and the PLRA

In the Court’s prisoners’ rights cases, deference is routinely mobilized—usually explicitly, although sometimes
implicitly76—to justify outcomes benefiting defendants at
the expense of plaintiffs. Yet in a related body of doctrine,
those cases involving the PLRA, one finds the Court
eschewing the deferential strategies it freely employs in the
prisoners’ rights cases. This is not to say these strategies
are not to be seen in the PLRA cases. One obvious example
of continuity in this regard can be found in Woodford v.
Ngo,77 in which the Court held that failure to properly
exhaust a prison’s grievance procedure, whatever the circumstances, constitutes a procedural default.78 After
Woodford, prisoners must “complete the [grievance] process in accordance with the applicable procedural rules,
including deadlines, as a precondition to bringing suit in
federal court.”79 Ngo argued that this holding would only
incentivize “prison administrators to devise procedural
requirements . . . designed to trap unwary prisoners and
thus to defeat their claims,” but the Court brushed this
concern aside, asserting instead that prison officials “concerned about maintaining order . . . have a reason for
creating and retaining grievance systems that provide . . . a
meaningful opportunity for prisoners to raise meritorious
grievances.”80 For those familiar with the actual dynamics
of carceral institutions, the Court’s reasoning seemed very
much at odds with the realities of prison life.81 And sure
enough, in his analysis of the prison grievance procedures
in several (but not all) states before and after Woodford,
Derek Borchardt found that, although the grievance processes in some states had been “reasonably updated,”82 in
several others, grievance procedures had been revised postWoodford “in ways that cannot be understood as anything
but attempts at blocking lawsuits.”83
In other of the Court’s PLRA cases, however, there is a
refusal to endorse at least some of the deferential strategies
that the Court readily adopts in its prisoners’ rights jurisprudence. In Jones v. Bock,84 for example, the Court
rebuffed efforts by the Sixth Circuit to impose additional
procedural burdens on prisoners bringing claims.85
Inspired by the PLRA’s exhaustion requirement86—itself a
statutory analogue to the deferential construction of substantive doctrine operating in cases like Turner, Whitley, and
Farmer—the Sixth Circuit adopted a number of procedural
rules, not present in the text of the statute, intended “to
implement this exhaustion requirement and facilitate early
judicial screening” of prisoner complaints.87 These rules:
(1) required a prisoner wishing to file a § 1983 action “to
allege and demonstrate exhaustion in his complaint,”
(2) “permit[ted] suit only against defendants who were
identified by the prisoner in his [initial] grievance,” and
(3) “requir[ed] courts to dismiss the entire action if the prisoner fail[ed] to [exhaust] any single claim in his complaint.”88

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 249

249

17/04/12 12:19 PM

As the Jones Court noted, these rules had no basis in either
the PLRA—which simply states that prisoners may not
bring suit in federal court “until such administrative remedies as are available are exhausted”89—or in the Federal
Rules of Civil Procedure.90 For example, as to the requirement that exhaustion be demonstrated in the plaintiff’s
initial pleading, under the Federal Rules, exhaustion is wellunderstood to be an affirmative defense.91 It must be
invoked by defendants in their response to the complaint,92
or else be deemed to have been waived; it is thus not the job
of the plaintiff to plead exhaustion in his or her initial filing.
Regarding the rule that a suit may not proceed unless the
named defendant was also named in the original grievance,
Jones noted that it is the grievance procedures themselves,
and not the PLRA, that establish the requirements of a
properly filed complaint.93 And finally, as to the rule that no
action may proceed unless all listed claims are exhausted,
Jones pointed to the standard judicial practice that “if a
complaint contains both good and bad claims, the court
proceeds with the good and leaves the bad.”94
Given the absence of statutory or doctrinal support for
the Sixth Circuit’s additional pro-defendant procedural
requirements, one might wonder how Jones could have
possibly come out other than it did. In each case, the rules
under scrutiny represented departures from standard
practice apparently invented out of whole cloth to benefit
defendants. But as has been seen, one can find in the
Supreme Court’s prisoners’ rights cases other examples
of well-established procedural rules being rewritten to bolster the position of prison officials at the expense of
prisoners.95 In Jones, however, the Court took a dim view
of such an undertaking. The Court credited the Sixth Circuit with good faith in the matter, assuming its motive in
imposing these rules was only to ensure that “the ‘new
regime’ mandated by the PLRA for prisoner complaints . . .
function[ed] effectively.”96 Yet it stated clearly that “courts
should generally not depart from the usual practice under
the Federal Rules on the basis of perceived policy concerns.”97 Indeed, the Court admonished the Sixth Circuit
that doing so would exceed the authority of the courts in
the constitutional scheme, quoting Justice Frankfurter to
the effect that “‘[w]hatever temptations the statesmanship
of policy-making might wisely suggest,’ the judge’s job is
to construe the statute—not make it better.”98 Nor, the
Jones Court emphasized, should the federal court read into
a statute a congressional intent to reframe a standard judicial practice; the norm in the federal courts was that,
where claims are “bad,” “only the bad claims are dismissed; the complaint as a whole is not,” and had
Congress “meant to depart from the norm,” the statute
would offer some indication to this effect.99
The Jones Court thus took the Sixth Circuit somewhat to
task for doing what the Court itself did in North Carolina
Prisoners’ Union and in Banks. Admittedly, there are differences between Jones on the one hand and North Carolina
Prisoners’ Union and Banks on the other. Jones involved statutory interpretation, and the statute itself provided the

250

FSR2404_04.indd 250

Court with a firm baseline from which to assess the Sixth
Circuit’s creative rule-making. By contrast, North Carolina
Prisoners’ Union and Banks involved constitutional claims,
as to which the Court sets its own baseline. In the statutory
context, it is up to Congress to weigh the various policy
concerns, whereas in the universe of constitutional claims,
that burden falls to the Court. But this distinction may
obscure as much as it reveals. For one thing, in Banks, the
Court as freely rewrote the Federal Rules as did the Sixth
Circuit in Jones; after Banks, the summary judgment rules
for prisoner cases are different, and more defendantfriendly, than in other cases. If this is not “depart[ing] from
the usual practice under the Federal Rules on the basis of
perceived policy concerns,”100 it is hard to know what
would be. As for the (re)description of the district court’s
role in evaluating evidence in prison cases in North Carolina Prisoners’ Union, Justice Rehnquist did not even try to
ground the change in any constitutional principles, besides
that of the imperative that courts “ordinarily defer to
[defendants’] judgment” in matters touching a prison’s
“security and order.”101 Yet as has been seen, that imperative can be readily invoked to justify any number of judicial
moves—the construction of doctrine, the recasting of
established procedural rules, the reframing of factual or
procedural history—that benefit defendant prison officials
at the expense of prisoner plaintiffs, and thus offers no
principled basis for distinguishing North Carolina Prisoners’ Union from Jones.102
And Jones is not the only PLRA case in which the Court
distances itself from deferential moves it employs elsewhere in its prison law cases. The Court’s PLRA
jurisprudence also betrays some inclination to reject a second form of deference found throughout its prisoners’
rights cases: reframing the facts in ways sympathetic to
defendant prison officials. Certainly, the PLRA cases still
show signs of this strategy, as the above discussion of
Woodford suggests. But in Brown v. Plata103—the biggest
and most consequential prison case in almost two
decades—the Court declined the invitation issued in Justice Alito’s dissent to interpret the facts in a way that,
although defendant-friendly, was stubbornly at odds with
the reality of the case. In doing so, the Court showed itself
able to acknowledge strategic behavior on the part of
prison officials and to credit the lived experience of people
in prison—capacities rarely on view in the prisoners’
rights cases.
The legal question presented in Plata was whether a
three-judge panel, struck pursuant to 18 U.S.C. § 3626,
exceeded its authority when it ordered California to reduce
its prison population to 137.5 percent of the rated capacity
of its facilities.104 Writing for the Court, Justice Kennedy
recognized that, although the order was “of unprecedented
sweep and extent,” the “medical and mental health care
provided by California’s prisons” had “for years . . . fallen
short of minimal constitutional requirements” and that
“[e]fforts to remedy the violation ha[d] been frustrated
by severe overcrowding. . . .”105 The majority opinion

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

rehearsed in detail some of the many disturbing facts of
the case, including that, due to “a shortage of treatment
beds, suicidal inmates may be held for prolonged periods
in telephone-booth sized cages without toilets”; that one
such inmate, having “been held in such a cage for nearly
24 hours,” was found “standing in a pool of his own urine,
unresponsive and nearly catatonic” because prison officials
had “no place to put him”;106 and that due to lack of space,
“up to 50 sick inmates may be held together in a 12-by
20-foot cage for up to five hours awaiting treatment.”107
Justice Kennedy then, as Linda Greenhouse aptly put it,
“plowed methodically through every section of the [PLRA]”
and agreed with the panel’s conclusion as to each.108
Noting that the Court’s stance was “necessarily deferential” as to the panel’s findings of fact, the majority
opinion canvassed some of the evidence heard at trial and
found it to support the conclusion that crowding was, as
required by § 3626, the “primary cause” of the constitutional violation.109 It also endorsed the panel’s finding that
“no other relief” would remedy the violation, agreeing that
although alternatives existed in theory, there was “no realistic possibility that California would be able to build itself
out of the crisis” and that any remedy requiring “significant additional spending by the state” was “chimerical.”110
As Justice Kennedy put it, “The Court cannot ignore the
political and fiscal reality behind this case”: the state legislature had been unable or unwilling “to allocate the
resources necessary to meet this crisis” and “[t]here is no
reason to believe it will begin to do so now,” when it faces
“an unprecedented budgetary shortfall.”111 Finally, the
Court found that the panel had given “substantial weight”
to public safety,112 noting the considerable statistical and
other testimony evidence heard at trial indicating that the
order “would have little or no impact on public safety.”113
Defending the panel’s conclusion as sufficiently supported, the Court observed that the PLRA “does not
require the court to certify that its order has no possible
adverse impact,” since such a reading “would depart from
the statute’s text by replacing the word ‘substantial’ with
‘conclusive.’”114
In sum, the three-judge panel in Plata had systematically analyzed its own proposed remedial order in light of
statutory requirements, and the Supreme Court, reviewing
that analysis in light of the facts and procedural history,
agreed that the panel had met the strict burdens of § 3626,
and thus upheld the order. Acknowledging the deference
it owed to the panel’s findings of fact, the Court credited
the panel’s account of the evidence and echoed the realism
that attended the panel’s deliberations as well as its reluctance to allow defendants’ implausible assertions to derail
statutorily authorized remediation.
Ordinarily there would be nothing surprising here;
appellate courts undertake this sort of review all the time.
But for a prison law case, Plata’s majority opinion was
notable for what it did not do, i.e., avail itself of any of the
deferential moves that more typically inform the Court’s
rulings in this area. It did not, for example, rewrite the

rules of appellate review, but forthrightly acknowledged
that “[w]ith respect to the three-judge court’s factual findings, this Court’s review is necessarily deferential.”115 In
assessing the panel’s conclusion that overcrowding was
the “primary cause” of the violation—as it needed to be if
the panel’s order was to satisfy § 3626—the Court even
passed up a chance to justify de novo review: although it
noted that the “issue of primary cause presents a mixed
question of law and fact,” it found that in this case “the
mix weighs heavily on the fact side.”116
It was left to Justice Alito in dissent to advocate the
recasting of rules governing appellate review. Although it
is standard practice for trial courts to assess expert testimony as part of its fact-finding role, Justice Alito insisted
that, given the issue presented, the weighing of expert evidence in this case was of a different order. As he put it,
when a trial court considers expert testimony “on broad
empirical questions such as the efficacy of preventing
crime through the incapacitation of convicted criminals,”
its “choice is very different from a classic finding of fact
and is not entitled to the same degree of deference on
appeal.”117 Justice Alito offered no justification for this
departure from the standard rules of appellate review
beyond this recitation of the substantive issue; indeed, he
seemed not even to realize such a justification was warranted. Perhaps more revealing still, Justice Alito’s dissent
made clear that no amount of evidence could convince
him that the panel had satisfied its obligation under
§ 3626(a)(1) to “give substantial weight to any adverse
impact on public safety.”118 In taking this position, Justice
Alito thus indicated not only that he would have departed
from accepted standards for appellate review, but also that
he would have effectively negated even the limited statutory protection of § 3626 in a case involving the largest
imaginable affected class.
But Justice Kennedy’s opinion in Plata is particularly
notable for its eschewal of the third form deference can
take in prison law cases, a stance evident in its willingness to take the record at face value and accept the facts as
given without seeking to reframe them in defendantfriendly ways. Again, it was left to Justice Alito in dissent
to present the facts in ways inconsistent with the record,
which he did most obviously in his discussion of the
PLRA’s narrow tailoring requirement.119 The three-judge
panel had found that without reducing prison overcrowding, the state would be unable to ensure the
constitutionality of its medical and mental health care. It
based this finding on California’s years of noncompliance with seventy-plus judicial orders issued in the two
cases.120 As the panel explained, it was not that the defendants were unaware of what reforms were needed, but
that the state had plainly demonstrated over two decades
its inability to effect them.121
Justice Alito’s discussion brushed this history aside.
Instead, it focused narrowly on what it would take to remedy the “deficiencies” in the prison’s medical and mental
health care delivery systems122 and reads as if its proposed

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 251

251

17/04/12 12:19 PM

solutions had simply escaped judicial notice. According to
Justice Alito:
Many of the problems [identified by the Court] could
be addressed without releasing prisoners and without
incurring the costs associated with a large-scale
prison construction program. Sanitary procedures
could be improved; sufficient supplies of medicine
and medical equipment could be purchased; an adequate system of records management could be
implemented; and the number of medical and other
staff positions could be increased. Similarly, it is hard
to believe that staffing vacancies cannot be reduced
or eliminated and that the qualifications of medical
personnel cannot be improved by any means short of
a massive prisoner release.123
Justice Alito thus concluded that “[w]ithout specific
findings backed by hard evidence, this Court should not
accept the counterintuitive proposition that these problems cannot be ameliorated by increasing salaries,
improving working conditions, and providing better training and monitoring of performance.”124 In the context of
this case, however, this reasoning is wholly at odds with
the facts. Taken together, the histories of Plata and Coleman (the two consolidated cases at issue) featured over
two decades of the state’s inability to implement the particular changes—including those listed by Justice Alito—that
all parties agreed were necessary to ensure constitutional
levels of care. The panel issued its order after years of
working with the state to try to bring about constitutional
compliance, finally concluding that no progress could be
expected absent a reduction in crowding. Yet Justice
Alito wrote as if the panel’s order was its first resort, and
as if the decades of failed efforts at compliance had
never occurred.
Thus, in Plata, at least two forms of deference typical of
prisoners’ rights cases were on view in Justice Alito’s dissent, while the majority eschewed them both in favor of a
systematic and even-handed analysis of the questions presented. Perhaps too much should not be made of this
departure from the Court’s usual treatment of prison law
issues; the case was so massive and the conditions so
extreme that there is unlikely ever to be another like it. Still,
it seems noteworthy that the two cases in which the Court
most obviously abstained from using the deferential moves
it employs elsewhere in the prison law context are PLRA
cases. With the passage of the PLRA, Congress enacted a
set of rules to govern prisoners’ rights cases that, as with
the standards established in Turner, Farmer, and Whitley,
directly manifest the imperative of judicial deference to
prison officials.125 Some of these same provisions, moreover, impose additional procedural burdens on prisoners
bringing § 1983 actions, creating exceptions to established
rules for the prisoners’ rights context much like the Court
did in North Carolina Prisoners’ Union and Banks.126 It may
only be a coincidence that Jones and Plata are both PLRA
cases. But it is also possible that even a Court ordinarily

252

FSR2404_04.indd 252

inclined to adopt deliberative strategies sympathetic to
defendant prison officials may hesitate to do so in cases
where the applicable law already features both highly deferential substantive standards and a host of procedural rules
rewritten to be more onerous for prisoners than for other
§ 1983 plaintiffs. Given the burdens the PLRA imposes on
prisoner suits, defendants in the PLRA cases that make it to
the Supreme Court will already have enjoyed the benefit of
multiple forms of deference.127 Perhaps even in the prison
law context there may be an implicit upper limit to how
much deference to defendant prison officials the Court is
willing to allow.
Of course, any such account must allow for possible
variation among the Justices. Justice Scalia, for one, implies
in his Plata dissent that at least in some cases, there are no
limits to his determination to find a way to rule against prisoners. Not one to mince words, Justice Scalia dismissed the
panel’s proceedings as “a judicial travesty” and its order as
“absurd.”128 Lest one think this assessment was based on
considered analysis of the facts in light of the relevant PLRA
provisions, Justice Scalia made clear that for his part, even
had the order satisfied the PLRA’s requirements to the letter, it would not have been enough for him. As he put it, “[o]
ne would think that, before allowing the decree of a federal
district court to release 46,000 convicted felons”—a misstatement of the order,129 it bears noting—“this Court
would bend every effort to read the law in such a way as to
avoid that outrageous result.”130
This language reinforces the strong impression that a
canvassing of the Court’s prison law cases already creates—that rather than applying the relevant law in good
faith to the facts of cases involving prisoners, at least some
Justices will “bend every effort to read the law” to favor the
defendants at the expense of the plaintiffs.131 And for those
inclined to such manipulation, the doctrine has featured a
ready mechanism: the imperative of judicial deference,
which in the absence of established principles for its application has proved sufficiently flexible to allow the
impression of good faith legal reasoning while stacking
the deck in defendants’ favor. It is perhaps this flexibility,
afforded by the absence of any principled account to guide
(and constrain) the invocation of deference in the prison
context, that offers the strongest evidence of the need for
such an account.
III.  Towards a Principled Theory of Deference

Even in prisoners’ rights cases, the Court does not always
find for defendants. In Cleavinger v. Saxner,132 for example,
prison officials sought absolute immunity from suit for
actions taken in their capacity as hearing officers in prison
disciplinary matters.133 The Court, however, rebuffed these
efforts. Recognizing the pressures hearing officers feel to
credit the testimony of fellow officers against that of
inmates,134 the Court rejected defendants’ claim that
“ample safeguards” existed to protect the due process
rights of prisoners at disciplinary hearings.135 It thus concluded that only qualified immunity and not absolute

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

immunity was appropriate.136 And in Turner v. Safley, the
Court struck down a Missouri regulation prohibiting prisoners from marrying without the approval of the
superintendent137 after it scrutinized the justifications
offered in defense of the rule and found them illogical.138
Nor is the Court’s invocation of deference in prison law
cases always misplaced. In Lewis v. Casey,139 for example,
Justice Scalia was right to chastise the district court for
imposing a minutely detailed order for the revamping of
the Arizona prisons’ law libraries that had been crafted
without consulting the state.140 As Justice Scalia observed,
the better approach would have been to “charge[] the
Department of Correction with the task of devising a Constitutionally sound program,” thereby “permitting prison
administrators to exercise wide discretion within the
bounds of constitutional requirements”141—an approach, it
bears noting, to which the Plata and Coleman courts scrupulously adhered throughout the history of both those cases.
The problem in the prisoners’ rights cases, and in the
prison law cases more generally, is thus not the use of deference per se, but the lack of any guiding principles for its
application. As a body, the prisoners’ rights cases currently stand for the implicit proposition that a greater
measure of deference is warranted when the plaintiffs are
prisoners, but to date, support for this position has been
more assertion than argument. Indeed, given the profound, real-life consequences of its prison law cases, it is
telling how little effort the Court has expended to justify in
a meaningful way its readiness to defer. Instead, its
defense of deference in the prison context has most often
rested on the contents of one oft-repeated paragraph, first
appearing in the reform era case142 of Procunier v. Martinez,143 in which the Court referred to the “Herculean
obstacles” prison officials face to the “effective discharge
of [their] duties,” and described “the problems of prisons
in America” as “complex and intractable” and “not readily
susceptible of resolution by [judicial] decree.”144 The Court
thus concluded that “courts are ill equipped to deal with
the increasingly urgent problems of prison administration
and reform.”145 Perhaps nothing so plainly indicates the
absence of efforts by the Court to defend its inclination to
defer in this context than the subsequent career of this
paragraph, which has been deployed to justify a deferential stance in the Court’s prison law cases at least fourteen
times since Martinez was decided.146
If the Court’s prison law jurisprudence is to be legitimate, a theory of judicial deference for the prison law
context is required, to address the questions of when deference is justified, what forms it may legitimately take,
which aspects of a court’s deliberations it may appropriately influence, and the proper limits on its use. The
present survey suggests that a focus on the prison law
cases would be a worthwhile place to begin crafting such a
theory. Relevant questions would include whether deference in practice takes any forms beyond those I have
identified (and indeed whether what I have described as
forms of deference are accurately characterized as such);

whether the cases suggest any patterns as to when and in
what forms deference is invoked; and whether those patterns reveal any principled bases for that invocation. As to
this last, Jones and Plata would be a fruitful starting point,
since in these cases, with little fanfare, the Court forbore
from deploying the deferential strategies so often seen in
the prisoners’ rights cases in particular. In Part II above,
I offered a theory as to why this might be: in light of the
extremely pro-defendant cast of the PLRA, at least some
Justices may resist taking still further steps to benefit
prison officials. Whether this theory is ultimately compelling and what it might offer in terms of principled limits
on the employment of additional pro-defendant deferential strategies in general are questions meriting further
consideration.
At the same time, efforts to develop a principled
account of judicial deference for the prison law context
should also reach beyond the cases, to consider at a more
abstract level when, if ever, it is appropriate for courts to
make the kinds of moves routinely seen in the Court’s
prisoners’ rights jurisprudence. Doing so will require
grappling directly with the various institutional concerns
that might explain the Court’s determination to defer in
this context. These concerns, all of which feature in that
oft-cited Martinez paragraph, have been staples of the federal courts’ prison law cases for over a century.147 Most
prominent among them is the notion of relative institutional competency, i.e., the view that prison officials and
not the courts have the expertise necessary to deal with the
“complex and intractable” problems of American prisons.148 But they also include more general concerns with
the demands of federalism149 and separation of powers,150
as well as with the possibility of inviting a deluge of frivolous prisoner complaints.151 
These are just the sorts of institutional concerns that,
as Lawrence Sager famously observed, may keep the federal courts from enforcing constitutional norms “to their
full conceptual limits.”152 Sager argued that “when the federal courts restrain themselves for reasons of competence
and institutional propriety . . . , it is incongruous to treat
the products of such restraint as authoritative determinations of constitutional substance”153—in other words, that
a court’s decision for institutional reasons to narrow the
scope of constitutional protections should not be mistaken
for a decision on substantive constitutional grounds.
Applied to the prison context, Sager’s point can be read as
an admonition to prison officials that, even if a federal
court declines to fully enforce prisoners’ constitutional
rights, the state is still constitutionally bound to honor
those rights in a meaningful way. However, without principles justifying the nature and scope of judicial deference
to prison officials, it is impossible to determine whether
courts have properly forborne from more capacious
enforcement of prisoners’ constitutional rights. And following Sager’s logic, to the extent that the deference
currently shown to prison officials proves on reflection to
be excessive, there may well be grounds for thinking that

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 253

253

17/04/12 12:19 PM

courts too have been failing to live up to the full measure
of their constitutional obligations.
The legal field I have been calling “prison law” is actually an amalgam of several areas of law, chief among
them administrative law and constitutional law. Issues of
judicial deference—or “levels of scrutiny” in the constitutional context—are central to both these areas, and any
inquiry into appropriate deference for the prison law context may thus also need to consider, and likely wrestle
with, the tensions in both administrative law and constitutional law between judicial power on the one hand and
executive and legislative power on the other. Of the two,
administrative law in general seems particularly relevant
to an inquiry into appropriate limits on deference. Prisons are run by government agencies, albeit agencies with
special status in the administrative law universe.154 Much
of the doctrine in the administrative law arena is concerned with the appropriate scope of judicial deference to
agency determinations.155 A set of questions thus naturally arises: In what ways does or should deference in the
prison law context differ from deference granted to other
administrative agencies? Ought prisons to be treated the
same as other agencies? To what extent should the
answer depend on the nature of the challenged regulation
or practice? On the process by which it was established?156
As for constitutional law, the scholarly literature affords a
rich discussion regarding the various levels of scrutiny
courts will bring to bear on constitutional claims.157 In
the prisoners’ rights context, there is an obvious tension
between the Turner test, which if not identical to rational
basis review is certainly a species of it, and the fact that in
many cases, it is prisoners’ fundamental rights—ordinarily afforded heightened scrutiny—that are at issue.
Consideration of the ways that constitutional law scholars, and the Justices themselves, understand and justify
the existing scheme of judicial scrutiny may thus prove
useful to crafting a principled approach to deference in
the prison context.158
This way of approaching the issue assumes that deference in the prison context may be understood at least to
some extent as a species of a general, cross-doctrinal legal
concept. To some readers, however, this notion may seem
to give the Court too much credit. Given its minimal
efforts to justify the extent to which it defers to prison officials, and given that at least two forms that deference takes
in this context—the refashioning of procedural rules to
disadvantage prisoners, and a persistent readiness to interpret the facts to favor defendants—seem at least on the
surface of questionable legitimacy, and given the Court’s
generally evident determination to find ways to allow
defendant prison officials to prevail, one might instead
regard the Court’s repeated invocation in its prison law
cases of the need to defer as just a makeweight: a convenient, authoritative-seeming basis on which to ensure that
the prisoners almost always lose. Seen in this light, it is
misguided to think that the Court’s use of “deference” in
other contexts will shed light on the use of this notion in

254

FSR2404_04.indd 254

the prison context; although the term is found elsewhere,
there in practice is no meaningful commonality.
There may be something to this skeptical view. Certainly, on the surface, there seems scant resemblance
between, for example, the kinds of burdens the court
imposes on administrative agencies before it will defer to
their expertise in the crafting of regulations pursuant to
their statutory authority and the broad grants of discretion
to prison officials embodied in the prisoners’ rights doctrine.159 It is, moreover, hard to square the Court’s
heightened scrutiny in cases involving fundamental constitutional rights in general with the extremely limited
review accorded the policies and practices burdening the
fundamental constitutional rights of prisoners.
Yet the fact that a trans-substantive investigation into
the forms of judicial deference is likely to reveal the prison
law context as an outlier and even as a site of systematic
judicial bias against a particular plaintiff class does not
diminish the potential value of the enterprise. To the contrary, exposing dissimilarity in the way judicial deference
is used in other doctrinal contexts as compared with its
use in the prison law cases would only reinforce the need
for a principled account of deference in the prison context.
Such a comparative exercise, moreover, is sure to provide
potentially useful guidance for the crafting of such an
account, drawn from instances in which the Court more
aggressively exercises its obligations of constitutional
enforcement and judicial review.
At the same time, moreover, efforts to make sense of
the use of judicial deference in the specific context of prisons may also prove fruitful in the other direction, yielding
insights of broader doctrinal relevance. For example,
exploring the possibility that the Court’s prison law jurisprudence is informed by an implicit animus toward the
affected class will invite engagement with the more overtly
normative issues raised in Carolene Products’ famous footnote four,160 including whether and to what extent the
political status of plaintiffs in general may justifiably influence judicial deliberations.161 And examining the forms
deference takes in the prison context may also offer a
typology with general application. In addition to those
forms of deference explored in Parts I and II, the prison
law cases also reveal a further, more subtle dimension of
judicial scrutiny that doubtless has analogues in many
other areas: the degree to which, as a deliberative matter—
or perhaps as a matter of judicial psychology—courts are
willing to press on the strength of defendants’ arguments.162 As was seen in Part I, the prison law cases are
full of examples in which the Court endorses the defendants’ version of the facts even when strong evidence
contradicts it, and accepts the defendants’ claimed justifications for the challenged policy or practice even in cases
where minimal pressure on the proffered reasoning would
reveal its illogic. In some cases—for example, North Carolina Prisoners’ Union—the Court is explicit about its
obligation to accept defendants’ claims absent “substantial
evidence . . . that the officials have exaggerated their

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

response.”163 In other cases—for example, Whitley v.
Albers—the Court adopts without explanation the defendants’ version of the facts, even when evidence in the
record strongly contradicts their version and the procedural posture demands viewing the evidence in the light
most favorable to the plaintiffs.164 Either way, as Justice
Stevens trenchantly observed in his opinion in Turner,
“[h]ow a court describes its standard of review when a
prison regulation infringes fundamental constitutional
rights often has far less consequence for the inmates then
the actual showing that the court demands of the State in
order to uphold the regulation.”165 Efforts to develop a
principled account of judicial deference in the prison law
context may thus also prompt consideration of the way
this less tangible but no less significant form of judicial
scrutiny operates elsewhere.
The quest for a defensible account of judicial deference
in the prison law context invites analysis on multiple
dimensions: doctrinal, institutional, comparative, normative, and even psychological. All told, given the complexity
of the carceral enterprise, judicial deference to prison officials may well be more readily justified than judicial
deference in other contexts. But even so, there remains an
urgent need for clearly articulated principles, since absent a
principled basis for deploying deference, the law governing
prisons and prisoners cannot be regarded as legitimate by
those it most affects—nor, arguably, does it deserve to be.

	4	

	5	

	6	
	7	
	8	

	9	
	10	
11	
	12	
	13	
	14	
	15	
	16	
	17	

	18	

Notes

19	

	1	

	20	

	2	

	3	

I thank Rachel Barkow, Jody Freeman, Giovanna Shay, and
especially Dick Fallon for helpful comments and conversation, and Erin Earl for outstanding research assistance.
See Cooper v. Pate, 378 U.S. 546 (1964) (holding for the first
time that state prisoners can sue in federal court under
§ 1983 for violation of their constitutional rights); see also
Wolff v. McDonnell, 418 U.S. 539, 555–56 (1974) (“There
is no iron curtain drawn between the Constitution and the
prisons of this country.”).
For example, Procunier v. Martinez, 416 U.S. 396 (1974),
decided at the height of what is widely regarded as the
“reform” era of prisoners’ rights litigation, see Malcolm M.
Feeley & Edward L. Rubin, Judicial Policy Making and the Modern State 39 (1998), plainly demonstrates this ambivalence.
Although Martinez was decided in favor of the prisoners, 416
U.S. at 398, Justice Powell took pains to emphasize that
prison officials face “Herculean obstacles to effective discharge of [their] duties,” id. at 414, and that “courts are ill
equipped to deal with the increasingly urgent problems of
prison administration and reform,” id. at 405. And if Martinez
still went on to insist that “a policy of judicial restraint cannot encompass any failure to take cognizance of valid
constitutional claims,” id., it was its emphasis on the judicial
obligation to defer that proved to be the lasting legacy of the
case. While Martinez’s key doctrinal move—emphasizing that
restricting prisoners’ First Amendment rights also restricts
the First Amendment rights of free people, thereby justifying
heightened scrutiny of restrictions on prisoners’ freedom of
expression—did not even last the Term, see Pell v. Procunier,
417 U.S. 817 (1974) (upholding a prohibition on media
access to particular prisoners without even noting the First
Amendment interest of the reporters involved when analyzing
the prisoners’ First Amendment claim), Justice Powell’s

	21	
	22	

	23	

language in Martinez emphasizing the need for judicial deference has been quoted frequently in subsequent cases to
justify finding for prison officials. For further discussion of
the lasting influence of this language from Martinez, see infra
Part III and note 146.
See Sharon Dolovich, Cruelty, Prison Conditions, and the Eighth
Amendment, 84 N.Y.U. L. Rev. 881, 961–62 & n.306 (2009)
(describing the role of deference in the history of prisoners’
rights litigation).
Pub. L. No. 104-134, §§ 801–810, 110 Stat. 1321–66 (1996)
(codified in scattered sections of 11, 18, 28, and 42 U.S.C.).
549 U.S. 199 (2007).
131 S. Ct. 1910 (2011).
The cases in this essay involve civil rights claims brought
under 42 U.S.C. § 1983 and Bivens v. Six Unknown Named
Agents of Federal Bureau of Narcotics, 403 U.S. 388 (1971).
Thus, references to “defendants” throughout refer to the
state prison officials sued by prisoner-plaintiffs for constitutional violations.
Lewis v. Casey, 518 U.S. 343, 361 (1996).
482 U.S. 78 (1987).
Id. at 89.
Id. (quoting Block v. Rutherford, 468 U.S. 576, 586 (1984)).
Id. at 90.
Id.
Id.
Id. at 89–90.
Id. at 90 (citation omitted) (quoting Jones v. N.C. Prisoners’
Labor Union, Inc., 433 U.S. 119, 131 (1977); Pell v. Procunier, 417 U.S. 817, 827 (1974)) (internal quotation marks
omitted). In Overton v. Bazzetta, the Court went further, saying that alternatives “need not be ideal,” but “need only be
available.” 539 U.S. 126, 135 (2003).
Turner, 482 U.S. at 90.
Id. at 91.
See Columbia Human Rights Review, A Jailhouse Lawyer’s
Manual 377 (9th ed. 2011), available at http://www3.law.
columbia.edu/hrlr/jlm/toc/ (“In most cases, a prison regulation survives the Turner test because it only requires that
the government’s explanation for the regulation be rational
(that it makes sense). This is not a very high standard for
the government to meet.”); Margo Schlanger, Inmate Litigation, 116 Harv. L. Rev. 1555, 1605–07 & n.139 (2003)
(examining the results of an empirical study of inmate
cases and explaining that “inmates are only very rarely successful in their federal civil rights actions,” id. at 1605, in
part because of “extremely defendant-friendly standards,”
including Turner, id. at 1605 n.139). And even when a prisoner-plaintiff does prevail on a Turner analysis, the victory is
frequently short-lived. See, e.g., Overton v. Bazzetta, 286
F.3d 311, 316–18, 323–24 (6th Cir. 2002) (holding on Turner
grounds that Michigan’s restrictions on visitations violated
prisoners’ constitutional rights), rev’d, 539 U.S. 126 (2003);
Gerber v. Hickman, 264 F.3d 882, 887–90 (9th Cir. 2001)
(holding on Turner grounds that plaintiff’s constitutional
rights were violated by defendants’ refusal to allow him to
arrange for the transport outside the prison of his own
semen, which he (and his wife) wished for his wife to use for
purposes of artificial insemination), rev’d, 291 F.3d 617
(9th Cir. 2002) (en banc).
475 U.S. 312 (1986).
E.g., Estelle v. Gamble, 429 U.S. 97, 104 (1976); see also Wilson v. Seiter, 501 U.S. 294, 303 (1991).
Whitley, 475 U.S. at 320–21 (quoting Johnson v. Glick, 481
F.2d 1028, 1033 (2d Cir. 1973) (Friendly, J.)) (internal quotation mark omitted). The Whitley standard initially applied only
to the use of force in a “disturbance.” Id. But in Hudson v.
McMillian, the Court extended Whitley to all use of force

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 255

255

17/04/12 12:19 PM

	24	
	25	

	26	
	27	
	28	

	29	
	30	
	31	

	32	
	33	
	34	
	35	
	36	

	37	
	38	

	39	

	40	
	41	
	42	
	43	

	44	
	45	

	46	
	47	
	48	

	49	
50	
	51	

	52	

	53	
	54	
	55	
	56	
	57	

	58	
	59	
	60	
	61	
	62	
	63	
	64	

256

FSR2404_04.indd 256

claims, whether arising from a “disturbance” or not. See 503
U.S. 1, 7 (1992).
511 U.S. 825 (1994).
See id. at 838 (“[A]n official’s failure to alleviate a significant
risk that he should have perceived but did not, while no cause
for commendation, cannot . . . be condemned as the infliction of punishment.”).
501 U.S. 294 (1991).
Id. at 300.
See Dolovich, supra note 4, at 895–907 (arguing this view in
detail).
433 U.S. 119 (1977).
See id. at 122–23.
N.C. Prisoners’ Labor Union, Inc. v. Jones, 409 F. Supp. 937,
942 (E.D.N.C. 1976).
Id. at 943.
Id. at 944.
Id.
Id.
Jones v. N.C. Prisoners’ Labor Union, Inc., 433 U.S. 119, 121
(1977).
Id. at 126.
Id. at 127–28 (emphasis added) (citations omitted) (internal
quotation marks omitted).
See also id. at 132 (“Appellant prison officials concluded that
the presence, perhaps even the objectives, of a prisoners’ labor
union would be detrimental to order and security in the prisons. It is enough to say that they have not been conclusively
shown to be wrong in this view.” (internal citations omitted)).
N.C. Prisoners’ Labor Union, 409 F. Supp. at 943.
548 U.S. 521 (2006).
Id. at 527–28, 536 (plurality opinion).
Id. at 529 (quoting Turner v. Safley, 482 U.S. 76, 89 (1987))
(internal quotation marks omitted); id. at 531.
Id. at 530.
Id. at 531 (alteration in original) (internal quotation mark
omitted).
See id. at 534.
Id.
Id. at 535 (quoting Banks v. Beard, 399 F.3d 134, 142 (3d
Cir. 2005)) (internal quotation marks omitted).
Id. at 543–52 (Stevens, J., dissenting).
Fed. R. Civ. P. 56(a).
Banks, 548 U.S. at 535 (plurality opinion). In his concurrence,
Justice Thomas, joined by Justice Scalia, reaffirmed his position that prisoners have no constitutional rights apart from
the Eighth Amendment. See id. at 537 (Thomas, J., concurring). There were thus arguably six votes supporting the
Banks plurality’s more restrictive summary judgment standard for cases brought by prisoners.
Masson v. New Yorker Magazine, Inc., 501 U.S. 496, 520
(1991).
Banks, 548 U.S. at 530 (plurality opinion) (emphasis added).
Id.
Id. at 554 (Ginsburg, J., dissenting).
Id. at 556 (emphasis in original).
Id. As Justice Ginsburg put it, “the plurality effectively tells
prison officials they will succeed in cases of this order, and
swiftly, while barely trying.” Id.
452 U.S. 337 (1981).
Id. at 346, 348.
Id. at 348.
Id. at 371 & n.3 (Marshall, J., dissenting).
See 475 U.S. 312, 317–18 (1986).
Id. at 322–23.
Id. at 330 (Marshall, J., dissenting); see also id. at 331
(“Although the Court sees fit to emphasize repeatedly ‘the
risks to the life of the hostage and the safety of inmates . . . ,’

	65	
	66	
	67	
	68	
	69	
	70	
	71	
	72	
	73	
	74	
	75	

	76	

	77	
	78	
	79	
	80	
	81	

	82	

	83	

I can only point out that respondent bitterly disputed that any
such risk to guards or inmates had persisted. The Court just
does not believe his story.” (internal citation omitted)).
539 U.S. 126 (2003).
Id. at 135.
Id.
518 U.S. 343 (1996).
430 U.S. 817 (1977).
Id. at 828.
Casey, 518 U.S. at 354 (citation omitted).
Id. at 354.
Bounds, 430 U.S. at 825.
Id. at 826.
Of course, depriving prisoners of outside visits, one of the
few bright spots in an otherwise dispiriting existence, may
well simply push those prisoners inclined to drugs or alcohol
to use these substances as a way to dull their pain.
In almost all the cases discussed above, the Court justifies
the moves it makes by explicit reference to the need to defer.
See infra note 146 & accompanying text; see also Beard v.
Banks, 548 U.S. 521, 530 (2006) (plurality opinion) (justifying the change in the summary judgment standard as applied
to prisoners by the Court’s obligation to defer to prison officials on “matters of professional judgment”); Whitley v.
Albers, 475 U.S. 312, 321–22 (1986) (citing language in
Rhodes v. Chapman, 452 U.S. 337, 349 (1981), and Bell v.
Wolfish, 441 U.S. 520, 547 (1979), regarding the need for
judicial deference to prison officials). In cases where it does
not, the imperative to defer is generally already embodied in
the substantive standards being applied—as in Overton v.
Bazzetta, 539 U.S. 126 (2003), which applied the Turner standard to regulations restricting plaintiffs’ visiting privileges.
The one exception is Farmer v. Brennan, 511 U.S. 825 (1994),
which does not explicitly invoke the need to defer in crafting
the standards for Eighth Amendment prison conditions
claims. In that case, however, it is hard to understand either
the Court’s holding or the justifications offered for it as anything other than the determination to create a zone for
official action in which prison officials can exercise discretion
without fear of constitutional liability. For further discussion
of the deferential character of the Farmer standard, see
Dolovich, supra note 4, at 895–907, 940–72.
548 U.S. 81 (2006).
Id. at 87.
Id. at 88.
Id. at 102.
See Margo Schlanger & Giovanna Shay, Preserving the Rule of
Law in America’s Jails and Prisons: The Case for Amending the
Prison Litigation Reform Act, 11 U. Pa. J. Const. L. 139, 149–50
(2008) (“[T]he PLRA’s exhaustion rule actually provides an
incentive to [prison and jail] administrators . . . to fashion ever
higher procedural hurdles in their grievance processes. . . . Can
anyone reasonably expect a governmental agency to resist this
kind of incentive to avoid merits consideration of grievances?”).
Derek Borchardt, Note, The Iron Curtain Redrawn Between
Prisoners and the Constitution, 43 Colum. Hum. Rts. L. Rev.
(forthcoming 2012) (manuscript at 58–59) (on file with
author) (listing Connecticut, Hawaii, Idaho, Indiana, and Mississippi as states with grievance procedures that seemed
“reasonably updated” after Woodford). But see id. (manuscript
at 29) (noting that because his research relied on states
being willing to share the details of their grievance procedures, “it is possible that the states that treat their grievance
procedures the most mischievously escaped scrutiny in this
Note, while more reasonable states received criticism due to
their willingness to make documents available”).
Id. (manuscript at 5); see id. (manuscript at 32–58) (listing
Alabama, Arizona, Arkansas, Colorado, Delaware, and

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

Oklahoma as states that either provide no guidance at all
for prisoners wishing to file grievances or that have revised
their requirements in ways “impossible to understand . . .
as anything but ‘procedural requirements that are designed
to trap unwary prisoners and . . . to defeat their claims,’” id.
(manuscript at 46) (second alteration in original) (quoting
Woodford, 548 U.S. at 102)).
		
 Among other strategies adopted post-Woodford, state
Departments of Correction reduced from ten days to five the
length of time available for filing or appealing a grievance
(Arizona), id. (manuscript at 37); revised the rules to require
that the initial grievance “specifically name” the personnel
involved, thereby justifying dismissal at later stages if the
grieving inmate initially misidentified or omitted the names
of responsible officials (Arkansas), id. (manuscript at 42–44);
revised the grievance forms to omit description of procedural
requirements (Arkansas again), id. (manuscript at 48); mandated that the grievance process shall be deemed not to have
been exhausted if a filed grievance is found to be “incomplete, inconsistent with a former step, incomprehensible,
illegible . . . fails to request relief,” or “in any other way fails
to comply with” the grievance procedure (Colorado), (manuscript at 50–51); added additional steps to the grievance
process (Oklahoma), id. (manuscript at 55–57); and changed
the rules to require dismissal of the grievance if the inmate
alters the grievance forms or “attach[es] additional documents to the formal grievance, other than the requisite
forms” (Oklahoma), id. (manuscript at 57).
	84	 549 U.S. 199 (2006).
	85	 Id. at 202–03.
	86	 42 U.S.C. § 1997e(a) (2006).
	87	 Jones, 549 U.S. at 202–03.
	88	 Id. at 203.
	89	 42 U.S.C. § 1997e(a).
	90	 See Jones, 549 U.S. at 212.
	91	 Id.
	92	 See Fed. R. Civ. P. 8(c); Jones, 549 U.S. at 212–17.
	93	 See Jones, 549 U.S. at 218.
	94	 Id. at 221.
	95	 See supra Part I (discussing Jones v. N.C. Prisoners’ Union and
Beard v. Banks).
	96	 Jones, 549 U.S. at 213.
	97	 Id. at 212.
	98	 Id. at 216 (quoting Felix Frankfurter, Some Reflections on the
Reading of Statutes, 47 Colum. L. Rev. 527, 533 (1947)).
	99	 Id. at 221 (quoting Robinson v. Page, 170 F.3d 747, 748–49 (7th
Cir. 1999)) (alteration and internal quotation mark omitted).
	100	 Id. at 212.
	101	 Jones v. N. C. Prisoners’ Union, Inc., 433 U.S. 119, 128 (1977).
	102	 Comparing Jones with North Carolina Prisoners’ Union and
Banks makes clear the need for principled limits on the
deployment of deference in the prison law context in general,
whether the cases are statutory or constitutional. As Jones
itself intimated, if the system is to maintain its legitimacy,
these limits must be crafted with more than just policy preferences in mind.
	103	 131 S. Ct. 1910 (2011).
	104	 Id. at 1922–23.
	105	 Id. at 1923.
	106	 Id. at 1924.
	107	 Id. at 1925.
	108	 Linda Greenhouse, A Voice from the Past, N.Y. Times Opinionator (June 1, 2011 9:00 PM), http://opinionator.blogs.
nytimes.com/2011/06/01/a-voice-from-the-past/.
	109	 Plata, 131 S. Ct. at 1932–34.
	110	 Id. at 1938 (internal quotation marks omitted).
	111	 Id. at 1939. As the Court noted, “[T]he common thread connecting the State’s proposed remedial efforts is that they

would require the State to expend large amounts of money
absent a reduction in overcrowding.” Id.
	112	
Id. at 1941.
	113	
Id. at 1943.
	114	
Id. at 1941.
	115	
Id. at 1932.
	116	
Id. (quoting Lilly v. Virginia, 527 U.S. 116, 148 (1999)
(Rehnquist, C.J., concurring in the judgment)) (internal quotation marks omitted).
	117	
Id. at 1966 (Alito, J., dissenting).
	118	
See id. at 1965–67 (discussing a court’s obligation under the
PLRA to “give substantial weight to any adverse impact on
public safety or the operation of a criminal justice system
caused by the relief,” id. at 1965 (quoting 18 U.S.C.
§ 3626(a)(1)(A) (2006)).
	119	
Under the PLRA, federal courts may only issue remedial
orders in prison conditions cases if that relief “is narrowly
drawn, extends no further than necessary to correct the violation of the Federal right, and is the least intrusive means
necessary to correct the violation,” and if in crafting that
relief, the court “give[s] substantial weight to any adverse
impact on public safety . . . .” 18 U.S.C. § 3626(a)(1)(A).
	120	
Plata, 131 S. Ct. at 1930–31.
	121	
See Coleman v. Schwarzenegger, No. CIV S-90-0520 LKK JFM
P, at 102–14 (E.D. Cal. Aug. 8, 2009).
	122	
Plata, 131 S. Ct. at 1963 (Alito, J., dissenting) (noting,
among other deficiencies identified by the panel, exam
tables and counters “not routinely disinfected,” medical
facilities “in an abysmal state of repair,” necessary medications and medical equipment “unavailable,” low hiring
standards and “high vacancy rates” (internal citations and
quotation marks omitted)).
	123	 Id. at 1964.
	124	 Id.
	125	 See, e.g., 18 U.S.C. § 3626(e)(2) (2006) (requiring that “any
motion to modify or terminate” a consent decree regarding
prison conditions be treated as an “[a]utomatic stay”); 42
U.S.C. § 1997e(e) (2006) (stipulating that no prisoner can
bring a claim for “mental or emotional injury suffered while
in custody without a prior showing of physical injury”).
	126	 See, e.g., 28 U.S.C. § 1915(b)(1) (2006) (stipulating that,
although in the ordinary case, people who affirm that they
are indigent may file a complaint without paying the filing
fee, see 28 U.S.C. § 1915(a)(1), a prisoner who “brings a civil
action or files an appeal in forma pauperis . . . shall be
required to pay the full amount of a filing fee,” albeit over
time); 28 U.S.C. § 1915(g) (stipulating that any prisoner who
has had three previous claims or appeals “dismissed on the
grounds that [they were] frivolous, malicious, or fail[ed] to
state a claim upon which relief may be granted” is barred
from bringing any further claims in federal court “unless the
prisoner is under imminent danger of serious physical
injury”); 42 U.S.C. § 1997e(a) (stipulating that no prisoner
can bring a § 1983 action “until such administrative remedies as are available are exhausted”).
	127	
This was certainly the case in Jones, in which the appeals
court imposed added procedural burdens on top of the
PLRA’s already onerous exhaustion requirement. See supra
text accompanying notes 84–94. And the same was also true
in Plata, where the ongoing efforts of the district courts in
Plata and Coleman to give the state sufficient opportunity to
remedy acknowledged constitutional violations may well have
helped to make conditions in the California prisons as dire as
they had become by the time the Court heard the case. See
supra text accompanying notes 109–11, 120–21.
	128	 Plata, 131 S. Ct. at 1951 (Scalia, J. dissenting).
	129	 Id. at 1950. The panel’s order required the state to reduce the
population of its prisons to 137.5 percent of rated capacity,

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 257

257

17/04/12 12:19 PM

but it made clear that it was up to the state to decide how to
achieve that target; if it could build more prisons, for example, the state could close in on the required capacity without
releasing prisoners. See Coleman v. Schwarzenegger, No. CIV
S-90-0520 LKK JFM P, at 32–33 (E.D. Cal. Aug. 8, 2009). As it
happens, the state has opted for a different strategy, one that
will reduce the prisoner population without either releasing
people or building more prisons: what it calls “realignment,”
which entails requiring the county jails to shoulder more of
the burden for low-level offenders and parole violators. See
Cal. Assembly Bill No. 109 (2011); see also Governor of the
State of Cal., Governor Brown Signs Legislation to Improve Public Safety and Empower Local Law Enforcement (Apr. 5, 2011),
http://gov.ca.gov/news.php?id=16964.
	130	
Plata, 131 S. Ct. at 1950 (Scalia, J., dissenting).
	131	
In this regard, it seems noteworthy that Plata was written by
Justice Kennedy. In recent years, Justice Kennedy has joined
the progressive wing of the Court—and written the majority
opinion—in cases on key sentencing issues, including striking down the death penalty for defendants who committed
non-homicide crimes, see Kennedy v. Louisiana, 554 U.S.
407 (2008), or who committed their crimes as juveniles, see
Roper v. Simmons, 543 U.S. 551 (2005), and finding life
without parole unconstitutional when imposed on juveniles
for non-homicide crimes, see Graham v. Florida, 130 S. Ct.
2011 (2010). And even more to the point, Justice Kennedy
appears to have taken a personal interest in the American
penal system, speaking publicly against excessive sentences, the folly of mass incarceration, and the
overcrowding of the California prison system. See, e.g.,
Carol J. Williams, Justice Kennedy Laments the State of Prisons in California, U.S., L.A. Times, Feb. 4, 2010 (reporting on
a public address by Justice Kennedy at Pepperdine Law
School, in which he “criticized California sentencing policies
and crowded prisons” and “expressed obvious dismay over
the state of corrections and rehabilitation in the country”).
It is hard not to think that Justice Kennedy’s growing personal awareness on these issues informed his thinking on
this case, and led him to take at face value the facts presented in Plata.
	132	 474 U.S. 193 (1985).
	133	 Id. at 198–99 & n.5.
	134	 See id. at 204 (observing that hearing officers are “prison
officers . . . temporarily diverted from their usual duties,”
who “work with the fellow employee who lodges the charge
against the inmate upon whom they sit in judgment” which
means that they are “under obvious pressure to resolve a disciplinary dispute in favor of the institution and their fellow
employee”).
	135	
Id. at 206.
	136	
Id. at 202–06.
	137	
482 U.S. 78, 98–100 (1987).
	138	
See id. at 97–98 (finding the marriage regulation to be “an
exaggerated response to [the claimed] security objectives”
and rejecting the “love triangles” justification as lacking any
“logical connection” with the regulation, id. at 98).
	139	 518 U.S. 343 (1996).
	140	 See id. at 362–63. According to Justice Scalia, the state
ought to have been given the opportunity in the first instance
to develop a plan for bringing its prison law libraries into constitutional compliance. See id. at 363 (noting that the court,
“[h]aving found a violation of the right of access to the
courts,” should not have “conferred upon its special master, . . . rather than upon [state corrections] officials, the
responsibility for devising a remedial plan”).
	141	 Id. at 362–63 (quoting Bounds v. Smith, 430 U.S. 817,
818–19, 832–33 (1977)) (internal citations and quotation
marks omitted).

258

FSR2404_04.indd 258

	142	

See Feeley & Rubin, supra note 3, at 39 (identifying the years
1965 through 1986 as the reform era of federal prisoners’
rights).
	143	
416 U.S. 396 (1974). For further discussion of the “reform
era” aspects of Martinez, see supra note 3.
	144	
Procunier, 416 U.S. at 404–05.
	145	
Id. at 405.
	146	
See Brown v. Plata, 131 S. Ct. 1910, 1956 (2011) (Scalia, J.,
dissenting); Johnson v. California, 543 U.S. 499, 529–30
(2005) (Thomas, J., dissenting); McKune v. Lile, 536 U.S. 24,
37 (2002); Shaw v. Murphy, 532 U.S. 223, 229 (2001); Lewis
v. Casey, 518 U.S. 343, 386–87 (1996); Thornburgh v. Abbott,
490 U.S. 401, 407–08 (1989); O’Lone v. Shabazz, 482 U.S.
342, 349–50 (1987); Turner v. Safley, 482 U.S. 78, 84–85
(1987); Rhodes v. Chapman, 452 U.S. 337, 351 n.16 (1981);
Bell v. Wolfish, 441 U.S. 520, 531, 547 (1979); Houchins v.
KQED, Inc., 438 U.S. 1, 8 (1978); Jones v. N.C. Prisoners’
Union, Inc., 433 U.S. 119, 126 (1977); Bounds v. Smith, 430
U.S. 817, 835 (1977) (Burger, C.J., dissenting); see also Pell
v. Procunier, 417 U.S. 817, 826 (1974) (echoing the language
from Martinez in a case decided in the same Supreme Court
term). I thank Scott Dewey for his help in identifying the
instances in which the Court subsequently relied on this portion of Martinez.
	147	
See Feeley & Rubin, supra note 3, at 30–34 (discussing the
institutional justifications routinely invoked by the federal
courts in prison law cases during the “hands-off” era).
	148	 Procunier, 416 U.S. at 404–05.
	149	 See id. at 405 (“[W]here state penal institutions are involved,
federal courts have a further reason for deference to the
appropriate prison authorities.”).
	150	 See id. (“[M]ost [prison problems] require expertise, comprehensive planning, and the commitment of resources, all of
which are peculiarly within the province of the legislature and
executive branches of government.”).
	151	 See id. at 405 n.9 (“T]he capacity of our criminal justice system to deal fairly and fully with legitimate claims will be
impaired by a burgeoning increase of frivolous prisoner
complaints.”).
	152	 Lawrence Gene Sager, Fair Measure: The Legal Status of Underenforced Constitutional Norms, 91 Harv. L. Rev. 1212, 1221
(1978).
	153	 Id. at 1226.
	154	 See, e.g., Giovanna Shay, Ad Law Incarcerated, 14 Berkeley J.
Crim. L. 329, 344–51 (2009) (explaining that many state
departments of correction are exempted from some or all
of the requirements of their state’s administrative procedure acts).
	155	 See, e.g., Chevron, U.S.A., Inc. v. Natural Res. Def. Council,
Inc., 467 U.S. 837, 842–43 (1984) (holding that when a court
reviews an agency’s interpretation of a statute it administers,
“[i]f the statute is silent or ambiguous with respect to the
specific issue, the question for the court is whether the agency’s answer is based on a permissible construction of the
statute,” id. at 843); Motor Vehicle Mfrs. Ass’n of the U.S. v.
State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 40–57 (1983)
(conducting a searching review under the Administrative Procedure Act’s (APA) “arbitrary and capricious” standard of the
Department of Transportation’s decision to rescind an automobile safety regulation); Citizens to Pres. Overton Park, Inc.
v. Volpe, 401 U.S. 402, 410–20 (1971) (considering what
standard of review to apply under the APA to the Secretary of
Transportation’s decision to route a highway through a park,
and concluding that although “the Secretary’s decision is
entitled to a presumption of regularity,” that presumption
should not “shield his action from a thorough, probing, indepth review,” id. at 415); see also Stephen Breyer, Judicial
Review of Questions of Law and Policy, 38 Admin. L. Rev. 363,

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

17/04/12 12:19 PM

364–65 (1986) (noting that administrative law doctrine
“urges courts to defer to administrative interpretations of
regulatory statutes, while also urging them to review agency
decisions of regulatory policy strictly”).
	156	
As Giovanna Shay incisively notes, courts applying the Turner
test do not distinguish “among types of prison and jail
regulations, policies, and rules, creating a one-size-fits-all
approach.” Shay, supra note 154, at 341. Consequently, “corrections regulations promulgated pursuant to notice-andcomment rulemaking” and “far more informal policies or
practices” receive the same (generous) measure of deference
afforded by Turner. Id. This practice presents a striking contrast with the normal rules of administrative law, which
“indicate[] that courts should defer to agency interpretations
that were issued after notice-and-comment rulemaking proceedings and to those that were embodied in adjudications,”
and that such deference “might be required in other situations,” but otherwise “judicial deference to agency
interpretations [must] be earned by their persuasive power.”
Daniel J. Gifford, The Emerging Outlines of a Revised Chevron
Doctrine: Congressional Intent, Judicial Judgment, and Administrative Autonomy, 59 Admin. L. Rev. 783, 804 (2007) (describing
United States v. Mead Corp., 533 U.S. 218 (2001)).
	157	
See, e.g., Michael C. Dorf, Incidental Burdens on Fundamental
Rights, 109 Harv. L. Rev. 1175 (1996) (analyzing the Court’s
use of varyingly strict standards of review when considering
the constitutionality of incidental burdens on fundamental
rights — specifically, free speech, free exercise of religion,
and privacy rights); Suzanne B. Goldberg, Equality Without
Tiers, 77 S. Cal. L. Rev. 481 (2004) (describing the Court’s
three-tier Equal Protection framework and proposing a single-standard alternative); Gerald Gunther, The Supreme Court,
1971 Term—Foreword: In Search of Evolving Doctrine on a
Changing Court: A Model for a Newer Equal Protection, 86
Harv. L. Rev. 1 (1972) (arguing in favor of shifting from the

three-tier Equal Protection framework to one requiring “that
legislative means must substantially further legislative ends,”
id. at 20), Kenji Yoshino, The New Equal Protection, 124 Harv.
L. Rev. 747, 755–63 (2011) (describing the “framework of
tiered scrutiny,” id. at 755, that the Court employs in its
Equal Protection cases).
	158	
Conversely, looking at the existing constitutional scheme
through the lens of prison law may reveal inconsistencies
indicating that the broader framework needs rethinking.
Either way, the enterprise is sure to be a fruitful one.
	159	
See supra notes 154–155.
	160	
See United States v. Carolene Prods. Co., 304 U.S. 144, 152
n.4 (1938).
	161	
One could readily imagine a defense of lesser constitutional
protections for those members of society who have broken the
criminal law and thus forfeited equal consideration. Certainly,
this view seems implicitly to inform Justice Thomas’s view of
the matter. See, e.g., Beard v. Banks, 548 U.S. 521, 536–37
(2006) (Thomas, J., concurring in the judgment); Overton v.
Bazzetta, 539 U.S. 126, 139 (2003) (Thomas, J., concurring in
the judgment). But the notion that constitutional protections
may be forfeited by individual actions—particularly by people
already subjected to criminal punishment as penalty for
crimes committed—is highly contentious, and to the extent
that it does help to explain the Court’s prisoners’ rights jurisprudence in general, it ought at the very least to be surfaced
and explicitly defended.
	162	 For a fascinating study of judicial psychology in the prison
law context, see Feeley & Rubin, supra note 3.
	163	 Jones v. N.C. Prisoners’ Union, Inc., 433 U.S. 119, 128
(1977) (quoting Pell v. Procunier, 417 U.S. 817, 827 (1974)).
	164	 See Whitley v. Albers, 475 U.S. 312, 322–26 (1986); see also
text accompanying notes 62–64.
	165	 Turner v. Safley, 482 U.S. 78, 100 (1987) (Stevens, J., concurring in part and dissenting in part).

F e d e r a l S e n t e n c i n g R e p o r t e r   •   V o l . 2 4 , N o. 4   •  A p r i l 2 0 1 2

FSR2404_04.indd 259

259

17/04/12 12:19 PM