PRISON LEGAL NEWS, a non-profit, Washington Corporation, Plaintiff–Appellant, v. Brad LIVINGSTON, in his official capacity as the executive director of the Texas Department of Criminal Justice; Jennifer Smith; Ramona Creek, in her individual capacity; Sue Weeks, in her individual capacity, Defendants–Appellees.
-- June 01, 2012
Prison Legal News (“PLN”), a non-profit corporation, brought this § 1983 action against the Texas Department of Criminal Justice (“TDCJ”) and three individual TDCJ officials (with TDCJ, the “Defendants”) for alleged First Amendment and due process violations in connection with TDCJ's censorship of five books PLN wishes to distribute to inmates in the Texas prison system. The district court granted summary judgment to the Defendants, and we now AFFIRM.
FACTS AND PROCEEDINGS
1. TDCJ's Book–Review Procedures
As part of a comprehensive system of processing mail addressed to prisoners, mail-room employees at each of TDCJ's 96 prison units review incoming books for content that may cause problems in TDCJ facilities. Individual defendants Ramona Creek and Sue Weeks are mail-room workers. When a book arrives, the mail-room employees first check to see whether the title is listed in the TDCJ database. The database includes books that have previously been sent to TDCJ inmates and reviewed by TDCJ personnel. If the book is in the database, it is shown as having been either approved or denied, and the mail-room staff follow the existing designation without conducting a new review.
If the book is not in the database, the mail-room employee adds it to the database and conducts a review for content deemed inappropriate for the prison population. TDCJ's “Uniform Offender Correspondence Rules” (the “Policy”) set forth the following content-based reasons that a book should be rejected:
1) It contains contraband that cannot be removed;
2) It contains information regarding the manufacture of explosives, weapons, or drugs;
3) It contains material a reasonable person would construe as written solely for the purpose of communicating information designed to achieve the breakdown of prisons through offender disruption such as strikes, riots, or [security threat group] activity;
4) A specific determination has been made the publication is detrimental to offenders' rehabilitation because it would encourage deviant criminal sexual behavior;
5) It contains material on the setting up and operation of criminal schemes ․;
6) It contains sexually explicit images.
The Policy also prohibits rejecting books for certain content. For example, the Policy provides that TDCJ should not reject a book solely because it advocates legitimate use of the prison grievance system or criticizes prison authorities. The parties agree that the Policy, which is a revised version of the rules established in a consent decree that the district court approved in 1983 and vacated in 2002, is facially constitutional. TDCJ has no other written guidelines for censoring books.
Purportedly applying the Policy, the mail-room employee reviewing an incoming book makes an initial decision whether the book should be approved or denied. At present, over 92,000 books are in the database, and approximately 80,000 of those are approved. If the reviewing employee approves the book, he delivers it to the addressee. TDCJ has no automatic review procedure for approvals, and no sender or inmate has a reason to request an appeal, so initial approval decisions by mail-room staff are, at least in practice, final. For this reason, mail-room workers are encouraged to “err on the side of caution” and deny books they believe might be problematic.
If mail-room staff find content they determine to be objectionable in a new book, they deny the book without reviewing the book's remaining content or attempting to weigh the objectionable content against permissible content. Staff record in the database that the book was “denied” and the reason for the denial, with citations to specific pages. Mail-room employees receive training and supervision on the Policy from the Mail System Coordinators Panel (“MSCP”), the body that coordinates the prison mail system. Individual defendant Jennifer Smith is the Program Director in charge of the MSCP. MSCP attempts to provide guidance to mail-room staff in complying with the First Amendment.
When a book is initially denied by mail-room staff, TDCJ sends a written notice to the prisoner and the book's sender1 that (1) states that the book has been denied for content, (2) identifies the page(s) containing the objectionable content and the reason it is objectionable, and (3) explains how to appeal the mail-room's decision. Either the prisoner or the sender may appeal the denial of the book.
An appeal of a new book is sent to TDCJ's Director's Review Committee
(“DRC”), a body of TDCJ administrators. DRC typically delegates the appeal to a member of the MSCP, the body headed by Smith. A single MSCP member reviews the book, purportedly applying the criteria listed in the Policy. Generally, the MSCP member will either approve or deny the book, and such a decision is final. If the MSCP member is uncertain whether the book should be denied, a second MSCP member reviews the book. If the two MSCP members do not agree, the book is referred to the DRC as a whole for final review. Like the mail-room employees, the MSCP/DRC “do[es] not attempt to determine whether the remainder of the book contains other content which is not in violation of policy or which would ‘outweigh’ [the objectionable content].”
During this appeals process, the MSCP/DRC reviews only the denial form and the book itself in reaching its decision. The “appellant” inmate or sender has no formal opportunity to participate in the appeals process, either at a hearing or by submitting written arguments. Essentially, the right to appeal is only the right to request additional review from TDCJ. The DRC sends the parties involved written notice of the decision. An approved book is delivered to the prisoner. The appeal and its result are noted in TDCJ's book database. A list of newly approved and denied books is posted in each unit's law library to inform prisoners of decisions from the current and previous month.
After a book's denial has been upheld by a final DRC/MSCP decision, TDCJ does not send a denial notice to senders of subsequent copies of the book. The mailroom sends notice to the prisoner addressee only for the purpose of seeking instructions on how to dispose of the denied book.
2. PLN's Books
PLN is a non-profit organization whose core mission is “informing prisoners about their rights.” PLN publishes Prison Legal News, a monthly magazine about prisoners' rights. TDCJ allows prisoners to subscribe to the magazine. PLN also distributes approximately 50 books, covering a variety of subject matters, that it believes are helpful to prisoners. Relevant to this appeal, TDCJ has excluded from its facilities five of the books PLN distributes: Prison Masculinities, Don Sabo, et al., eds., (2001); The Perpetual Prison Machine: How America Profits from Crime, Joel Dyer (1999); Lockdown America: Police and Prisons in the Age of Crisis, Christian Parenti (2000); Soledad Brother: The Prison Letters of George Jackson, George Jackson (1970); and Women Behind Bars: The Crisis of Women in the U.S. Prison System, Silja J.A. Talvi (2007).
Prison Masculinities is a collection of essays from various authors about prison conditions. A mail-room employee denied the book in 2001 because pages 128–31 “contain rape” and pages 194 and 222 “contain racial material.” The prisoner addressee appealed and the DRC upheld the denial. In 2009, a TDCJ prisoner ordered a copy of the book from PLN, and it was rejected as a matter of course because it was marked as denied in the TDCJ database.
The offending pages of Prison Masculinities describe incidents of prison rape in detail, including passages describing how the author was “raped and beaten by Blacks as a punishment for permitting [himself] to be raped by Whites,” assaulted by “a huge gang” that “took turns raping us anally and orally
simultaneously,” forced to perform fellatio, urinated upon during a rape, and forced to drink urine during a “mass rape.” The “racial material” that was cited in the denial decision included racial slurs and an account of a prison guard “calling [a prisoner] a ‘nigger’ as the other guards laugh[ed].”
According to PLN, The Perpetual Prison Machine is a “critique of the prison industry.” Mail-room staff denied the book in 2000 because the description of rape on page 45 “encourage[s] ․ deviant criminal sexual behavior.” The prisoner addressee appealed and the denial was upheld by the DRC. The offending page described two rapes, including details about being held down and subjected to anal and oral sex simultaneously.
In 2009, two prisoners ordered copies of The Perpetual Prison Machine from PLN. The mail-rooms denied the book for the reason listed in the database, described above, and, consistent with its normal procedures, sent notice to the prisoners but not to PLN.
Lockdown America is a history of American prison expansion, beginning in the 1960s. A mail-room employee reviewed and denied the book in 2000. The denial was appealed and upheld by the DRC. Lockdown America was denied because page 206 contains “racial material.” In discussing California's prison system, this page describes racist prison-guard organizations and racial tension between guards and prisoners. No TDCJ prisoner ordered Lockdown America from PLN within the time frame relevant to this suit.
Soledad Brother is a collection of letters by George Jackson, a member of the Black Panthers who was imprisoned in California's Soledad prison in the 1960s. According to California prison officials, the book is used by the “Black Guerilla Family” prison gang to convey its ideology to recruits. In re Furnace, 185 Cal.App.4th 649, 110 Cal.Rptr.3d 820, 823 (Cal.Ct.App.2010). A mail-room employee reviewed and denied the book in 2005. The denial was appealed and upheld by the DRC. The reason for the 2005 denial was that racial material appears in the introduction on page xxiii. One paragraph contains the words “whitey” and “honky.”
In July 2010, Smith removed Soledad Brother from the book database so that it was not listed as approved or denied, and testified that she would not have denied the book based solely on the page cited by the original denial decision. Soon thereafter, PLN sent a copy of Soledad Brother to a prisoner. A mail-room employee denied the book. Because Smith had removed the previous denial indicator from the database, the mail-room sent written notice of the denial to the prisoner and PLN, just as in the case of a book reviewed for the first time, but PLN did not receive the notice. The prisoner did not appeal. The reasons given for the 2010 denial were that certain pages contain racial slurs; and the entire book advocates the overthrow of prisons by riots and revolt.
Later in 2010, the publisher of Soledad Brother sent a copy of the book to another TDCJ prisoner. The mail-room denied the book for the reasons listed in the database, and because the 2010 denial had not yet been appealed and upheld by the DRC, TDCJ sent written notice to the prisoner and the publisher. The prisoner appealed the denial, which was upheld by the DRC. No TDCJ prisoner ordered Soledad Brother from PLN within the time frame relevant to this suit.
PLN describes Women Behind Bars as an “exposé” on the treatment of women in prisons. A mail-room employee reviewed and denied the book in 2008 because page 38 describes sex with a minor, which would “encourage deviant criminal sexual behavior.” Neither the prisoner nor the sender (amazon.com) appealed. Page 38 contains the sentence: “The dark secret of her life was that she had been forced to perform fellatio on her uncle when she was just four years old.” The prisoner addressee notified the book's author, Silja J.A. Talvi, who notified PLN, that the book had been denied. Thereafter, PLN sent an unsolicited copy of Women Behind Bars to the same prisoner as a means of “Testing for Censorship.” The mailroom denied the book for the reason listed in the book database.
Later in 2008, an unknown sender mailed a copy of Women Behind Bars to another TDCJ prisoner. The book was denied, the prisoner appealed, and the denial was upheld by the DRC/MSCP. Around the same time, PLN sent unsolicited copies of the book to numerous TDCJ prisoners to test for censorship. Silja's assistant asked PLN to send copies of Women Behind Bars to eight TDCJ prisoners “to see what happens.” PLN also sent unsolicited copies of Women Behind Bars to other TDCJ prisoners, asking them to send PLN copies of denial notices that they receive.
In July 2010, Smith re-reviewed Women Behind Bars and determined that it should be approved for prisoners, even though she had stated in an answer to an interrogatory only a few months earlier that “no one in my office would have felt any doubt about denying the book” because it “describes deviant sexual behavior with a minor, [and] it would have been standard practice for the MSCP staff member who inspected this book to deny it.” Following Smith's re-review, the book is listed as approved in TDCJ's database.
3. Procedural History
Based on the exclusion decisions described above, PLN filed suit against the Defendants pursuant to § 1983, alleging First Amendment and due process violations and seeking damages and an injunction requiring TDCJ to allow PLN's books into its facilities. The district court granted the Defendants' qualified-immunity motion on the due process claims against the individual defendants on the grounds that any right of subsequent senders to receive notice was not clearly established. The parties filed cross-motions for summary judgment on the remaining claims. The only factual issue specifically identified was whether PLN received notice and an opportunity to respond when TDCJ personnel denied a prisoner access to a copy of Soledad Brother sent by PLN. PLN noted, however, that this fact might not be material. The district court granted Defendants' motion and denied PLN's. The court dismissed PLN's claims regarding Lockdown America and Soledad Brother, which no prisoner had ordered from PLN, for lack of standing on the grounds that PLN has no First Amendment right to send books to prisoners who have not requested them.2
The district court then held that Defendants did not violate PLN's First
Amendment rights by denying prisoners access to the books Women Behind Bars, Prison Masculinities, and The Perpetual Prisoner Machine. Finally, the district court dismissed PLN's due-process claims against TDCJ as moot based on the recent change to TDCJ's written procedures providing notice and the opportunity to appeal to senders rather than editors or publishers. The court determined that this change satisfied any due process requirements. The district court entered a final judgment dismissing the case, and PLN timely appealed.
STANDARD OF REVIEW
We review a grant of summary judgment de novo, applying the same standards as the district court. Apache Corp. v. W & T Offshore, Inc., 626 F.3d 789, 793 (5th Cir.2010). Summary judgment is appropriate when “there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(a). There is no genuine issue for trial “[i]f the record, taken as a whole, could not lead a rational trier of fact to find for the non-moving party.” Kipps v. Caillier, 197 F.3d 765, 768 (5th Cir.1999). We view the evidence in the light most favorable to the nonmoving party. Carnaby v. City of Hous., 636 F.3d 183, 187 (5th Cir.2011).
I. First Amendment Claims
1. Unsolicited Books
The district court concluded that the First Amendment does not guarantee the right of outsiders to send unsolicited communications to prisoners. On that basis, the court determined that PLN lacks standing to bring claims related to TDCJ's censorship of Lockdown America, Soledad Brother, or Women Behind Bars because no prisoner had requested a copy of these titles from PLN prior to the filing of this lawsuit. TDCJ argues (1) that the district court was correct to deny PLN standing on this basis and (2) even if PLN has standing, its claims relative to these three titles should lose on the merits. TDCJ is incorrect on both points.
Even if the First Amendment did not guarantee PLN the right to send unsolicited mail to prisoners, that would not destroy PLN's standing to bring its claims. To have standing at the summary judgment stage, PLN must present evidence of specific facts that, if true, would demonstrate an injury in fact that is fairly traceable to the defendant's conduct and likely to be redressed by a favorable ruling. See Lujan v. Defenders of Wildlife, 504 U.S. 555, 561, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). TDCJ contends that the constitutional “injury in fact” requirement includes the additional requirement that a plaintiff's injury in fact must be “caused by the violation of legal right.” (quoting Lewis v. Casey, 518 U.S. 343, 353 n. 4, 116 S.Ct. 2174, 135 L.Ed.2d 606 (1996)). From this premise, TDCJ argues that PLN lacks standing to bring claims regarding the unsolicited books simply because the First Amendment does not, in its view, protect the right PLN seeks to vindicate. We first note that TDCJ's argument would, in every case, condition standing on a full and favorable determination on the merits of a plaintiff's claim. But a plaintiff need not prevail on the merits before he can establish his standing to sue. For that reason, TDCJ's argument is clearly incorrect.
Government interference with one's attempts to sell or distribute written material unquestionably satisfies Article III's injury-in-fact requirement. See, e.g., Bantam Books, Inc. v. Sullivan, 372 U.S. 58, 64 n. 6, 83 S.Ct. 631, 9 L.Ed.2d 584 (1963) (“The constitutional guarantee of freedom of the press embraces the circulation of books․”).3 PLN's interest in distributing books to TDCJ's inmates—which is precisely the type of interest at the core of First Amendment protections—is more than sufficient to support its standing to sue.
TDCJ cites Lewis v. Casey to support its argument that Article III standing requires an injury-in-fact that is caused by the violation of a legal right. TDCJ misunderstands the Supreme Court's standing discussion in Lewis. See 518 U.S. at 351–53 & n. 4. Lewis involved an inmate's claim that he was denied his constitutional right of access to the courts by the poor quality of his prison's legal library. In that context, the Supreme Court said that a deficient law library, though it might be an “injury in fact,” was not a “relevant injury in fact” because there is not, and the plaintiff did not even argue for, a free-standing right to access a legal library. The right to access legal materials derives from the right to access the courts, so a “relevant” injury in Lewis had to be one that was at least allegedly caused by being denied access to the courts, i.e., lost or impeded legal claims. Consequently, the plaintiff in Lewis lacked standing because he did not even allege an injury that resulted from the alleged legal violation. Cf. Allen v. Wright, 468 U.S. 737, 751, 104 S.Ct. 3315, 82 L.Ed.2d 556 (1984) (noting that standing requires a “plaintiff [to] allege personal injury fairly traceable to the defendant's allegedly unlawful conduct” (emphasis added)).
Unlike the plaintiff in Lewis, PLN suffered an injury—denial of the opportunity to communicate with certain inmates—that resulted directly from the government action it challenges as illegal. Aside from its misplaced reliance on Lewis, TDCJ presents no argument as to why this injury fails to meet Article III's injury-in-fact requirement, and it is well recognized that booksellers have standing to challenge censorship schemes. PLN's asserted injury is neither speculative nor abstract. Thus, PLN has standing to raise its First Amendment claims regarding the censorship of Lockdown America, Soledad Brother, and Women Behind Bars.
TDCJ's argument that the First Amendment does not protect the right of outsiders to send unsolicited mail to inmates is unpersuasive. It is well-established that “[p]rison walls do not form a barrier separating prison inmates from the protections of the Constitution, nor do they bar free citizens from exercising their own constitutional rights by reaching out to those on the ‘inside.’ ” Thornburgh v. Abbott, 490 U.S. 401, 407, 109 S.Ct. 1874, 104 L.Ed.2d 459 (1989) (citing Turner v. Safley, 482 U.S. 78, 84, 94–99, 107 S.Ct. 2254, 96 L.Ed.2d 64 (1987)) (alteration in original) (citation and internal quotation marks omitted). “[A] prison inmate retains those [constitutional] rights that are not inconsistent with his status as a prisoner or with the legitimate penological objectives of the corrections system.” Turner, 482 U.S. at 95 (second alteration in original) (internal quotation marks omitted). Specifically, prisoners and their correspondents enjoy the protections of the First Amendment except to the extent that prison regulations curtailing those protections are “reasonably related to legitimate penological interests.” Id. at 89. Receiving unsolicited mail is not, by itself, intrinsically “inconsistent with [one's] status as a prisoner” or an obvious threat to valid penological objectives, and TDCJ does not allege that it is. Thus, TDCJ's argument would carve away a large chunk of the First Amendment's protections even without so much as an assertion that those protections are in conflict with legitimate penological interests. TDCJ's analysis essentially ignores the framework provided in Turner and applied in Thornburgh: as a default, prisoners and their correspondents have all the constitutional rights of those outside the prison context unless those rights are curtailed by a reasonable prison regulation. Following Turner and Thornburgh, we decline to go beyond the limitations imposed by TDCJ's regulations and deny prisoners rights that TDCJ has not found it necessary to curtail.
The government's brief relies heavily on an illicit inference drawn from the Supreme Court's statement in Thornburgh that “there is no question that publishers who wish to communicate with those who, through subscription, willingly seek their point of view have a legitimate First Amendment interest in access to prisoners.” 490 U.S. at 408.4 From this sentence, TDCJ concludes that First Amendment protections must not extend to unsolicited communications. The quoted sentence simply cannot do the heavy lifting TDCJ asks of it. The qualified language is best explained as tracking the specific facts before the Court in Thornburgh. Recognizing one right as obvious hardly demonstrates that a related right does not exist. In any event, the inference TDCJ attempts to draw is directly contrary to the whole approach set forth in Turner and followed consistently since: prisoners and those “reaching out” to them enjoy generally-existing constitutional rights unless those rights interfere with prison regulations reasonably related to legitimate penological interests. It is therefore unsurprising that TDCJ's position was rejected by the only court of appeals decision cited by either party that squarely addresses the issue. Hrdlicka v. Reniff, 631 F.3d 1044, 1049 (9th Cir.2011) (“A First Amendment interest in distributing and receiving information does not depend on a recipient's prior request for that information.”).
TDCJ correctly argues that the rights of outsiders wishing to communicate with prisoners are correlative to the rights of prisoners and must be analyzed under the same standard. But TDCJ is wrong to conclude that outsiders' rights are therefore strictly dependent on a prior request from an inmate. Rather, the correlative nature of the rights indicates that, in the absence of a prison regulation to the contrary, prisoners have the right to receive, and outsiders have a correlative right to send, unsolicited communications. See Sorrels v. McKee, 290 F.3d 965, 970 (9th Cir.2002) (recognizing a prisoner's right to receive an unordered “gift publication”). The general right to receive unsolicited communications free from government interference is not only well-established, it is also quite valuable, a fact that is particularly apparent in the prison context: prisoners have an obvious interest, for example, in receiving unsolicited mailings from family members attempting to reconcile, ministries reaching out to convicts, and those attempting to offer legal assistance, because prisoners would often be practically unable to initiate such contact themselves. Cf. Thornburgh, 490 U.S. at 407 (recognizing the qualified rights of lawyers, journalists, families, and friends to access prisons “though the written word” as “undoubtedly” valid without conditioning such rights on prisoner solicitation).
For the above reasons, PLN has standing to bring claims related to the three books it sent without a prior request. We also reject TDCJ's argument that the First Amendment offers no protection for unsolicited communications to prisoners. Because TDCJ's policies do not generally treat unsolicited books differently from solicited ones, we do not address whether a policy of rejecting all unsolicited books would be defensible under the standards delineated in Turner and Thornburgh.
2. Turner-Thornburgh Analysis
“[A] prison inmate ‘retains those [constitutional] rights that are not inconsistent with his status as a prisoner․’ ” Turner, 482 U.S. at 95 (quoting Pell v. Procunier, 417 U.S. 817, 822, 94 S.Ct. 2800, 41 L.Ed.2d 495 (1974)) (second alteration in original). Accordingly, prisoners and their correspondents enjoy the protections of the First Amendment except to the extent that prison regulations curtailing those protections are “reasonably related to legitimate penological interests.” Thornburgh, 490 U.S. at 404 (quoting Turner, 482 U.S. at 89). In assessing the “reasonableness” of a prison regulation that infringes on First Amendment interests, a court must consider four factors: (1) whether the regulation is “rationally related” to a legitimate penological goal; (2) whether alternative means of exercising First Amendment rights remain open; (3) the impact that accommodating the asserted right will have on other prisoners and prison employees; and (4) whether there are easy and obvious alternative means of accommodating the asserted right. Id. at 414–18. “[R]ationality is the controlling factor,” Mayfield v. Tex. Dep't of Criminal Justice, 529 F.3d 599, 607 (5th Cir.2008), and the remaining factors are best understood as indicators of rationality. See Scott v. Miss. Dep't of Corr., 961 F.2d 77, 80–81 (5th Cir.1992).
The Supreme Court has also instructed that it is “important to inquire whether prison regulations restricting inmates' First Amendment rights operated in a neutral fashion.” Thornburgh, 490 U.S. at 415 (quoting Turner, 482 U.S. at 90). This neutrality requirement means that “the regulation or practice in question must further an important or substantial governmental interest unrelated to the suppression of expression.” Id. (quoting Procunier v. Martinez, 416 U.S. 396, 413, 94 S.Ct. 1800, 40 L.Ed.2d 224 (1974)). Prison regulations that “draw distinctions between publications solely on the basis of their potential implications for prison security” are facially neutral in the relevant sense. Id. at 415–16.
PLN bears the burden of showing that the regulations, as applied, are not reasonably related to legitimate penological objectives, see Overton v. Bazzetta, 539 U.S. 126, 132, 123 S.Ct. 2162, 156 L.Ed.2d 162 (2003), or that they are an “exaggerated response” to such concerns. Turner, 482 U.S. at 87. Moreover, courts must give “substantial” deference to prison officials' judgment. See, e.g., Overton, 539 U.S. at 132. TDCJ must do more, however, than merely show “a formalistic logical connection between [its censorship decisions] and a penological objective,” Beard v. Banks, 548 U.S. 521, 535, 126 S.Ct. 2572, 165 L.Ed.2d 697 (2006); it must “show[ ] a reasonable relation,” in light of the “importance of the rights [here] at issue.” Id. at 533. To be entitled to summary judgment, then, the record must be “sufficient to demonstrate that the Policy is a reasonable one.” Id.
PLN mistakenly (and extensively) relies on this court's decision in Guajardo v. Estelle, 580 F.2d 748 (5th Cir.1978). In analyzing the constitutionality of an earlier version of TDCJ's mail regulations, Guajardo applied the test set out in Martinez, 416 U.S. 396, 94 S.Ct. 1800, 40 L.Ed.2d 224. But Thornburgh partially overruled Martinez, limiting its application to “regulations concerning outgoing correspondence,” and adopted the Turner standard, which is more deferential to prison administrators. 490 U.S. at 413. Thus,
Thornburgh must be read as modifying ․ Guajardo ․ in regard to prison regulations or practices which deal with prisoner mail. That is, in determining the constitutional validity of prison practices that impinge upon a prisoner's rights with respect to mail, the appropriate inquiry is whether the practice is reasonably related to a legitimate penological interest.
Brewer v. Wilkinson, 3 F.3d 816, 824 (5th Cir.1993). PLN makes much of the fact that a post-Thornburgh decision of this court cited Guajardo. But in citing Guajardo, the court sought only to demonstrate that the challenged regulations were undoubtedly facially constitutional because they had been upheld under the Martinez standard, which was less deferential to prisons. PLN is plainly wrong that Guajardo is controlling precedent.
a. Reasonableness of Specific Exclusion Decisions
i. Prison Masculinities and The Perpetual Prison Machine
Both of these books contain graphic depictions of prison rape. Smith testified in her deposition that such descriptions constitute a threat to prison safety and security because prisoners could use the descriptions as templates to commit similar rapes. PLN argues that TDCJ has not presented evidence that a book containing depictions of sexual abuse has ever incited a copycat crime. But TDCJ is correct that it is not required to prove that the censored books would cause problems. PLN bears the burden to show that there is not a reasonable relationship between TDCJ's practice and legitimate penological objectives. See Overton, 539 U.S. at 132 (“The burden ․ is not on the State to prove the validity of prison regulations but on the prisoner to disprove it.”). Moreover, the Supreme Court has specifically acknowledged that prison policies may be legitimately based on prison administrators' reasonable assessment of potential dangers. See, e.g., Thornburgh, 490 U.S. at 407 (“[P]rison officials may well conclude that certain proposed interactions, though seemingly innocuous to laymen, have potentially significant implications for the order and security of the prison.”).
Pursuant to the inmate-education requirements of the Prison Rape Elimination Act of 2003 (“PREA”), TDCJ has its own program for educating prisoners about sexual abuse, so it does not, as PLN implies, deny inmates information about the existence of this problem. PLN argues that TDCJ's basis for denying these books is illegitimate because other books containing depictions of prison rape have been approved. PLN specifically contends that the rape scene in Stephen King's novella Rita Hayworth and the Shawshank Redemption, which appears in multiple literary compilations that have been approved by TDCJ, is just as graphic as the designated passages of Prison Masculinities and The Perpetual Prison Machine.
Aside from pointing to this alleged inconsistency, PLN has not produced specific evidence demonstrating that TDCJ's censorship of Prison Masculinities and The Perpetual Prison Machine is “arbitrary or irrational.” Freeman v. Tex. Dep't of Criminal Justice, 369 F.3d 854, 861 (5th Cir.2004) (quoting Turner, 482 U.S. at 90). PLN argues that it and the authors of these books have no intention to incite sexual abuse, but it has not presented evidence that the depictions of rape contained in the books pose no threat to safety or rehabilitation. Instead, PLN only pits its own assessment of the likely consequences of allowing the books into prisons against the assessment of the TDCJ officials. It is, of course, extremely difficult for courts to judge whose assessment is more likely correct, which highlights the importance of the deference that is accorded to prison administrators applying reasonable policies.
ii. Soledad Brother
TDCJ now excludes Soledad Brother because it contains racial slurs and “advocates the overthrow of prisons by riot and revolt.” According to California prison officials, the book is used by the “Black Guerilla Family” prison gang to convey its ideology to recruits. PLN denies that the book advocates the violent overthrow of prisons, but does not significantly elaborate on the book's viewpoint or contents. PLN acknowledges that the book and its author are controversial, and arguably concedes that the book can be read as promoting race riots, contending that “a ‘reasonable person’ ․ could not conclude Soledad Brother's sole intent is to encourage race riots.”
Though PLN disagrees with TDCJ's assessment of the book and its likely effects on a prison population, it has not presented evidence demonstrating that TDCJ's exclusion of the book is irrational. TDCJ's exclusion decision is therefore entitled to deference.
iii. Women Behind Bars
Because TDCJ no longer censors Women Behind Bars, the relevant question is whether its past exclusion of the book was reasonably related to TDCJ's legitimate penological interests. TDCJ censored the book because it recounts the sexual molestation of a young child, which Smith and Weeks testified could impair the rehabilitation of sex offenders or cause disruptive outbursts by prisoners who were similarly victimized. Despite this explanation, Smith has determined that Women Behind Bars should no longer be excluded. Presumably, the decision to approve the book indicates that Smith does not believe the once-censored passage actually poses a threat to TDCJ's legitimate goals, but this reassessment alone does not establish that the past exclusion was necessarily unreasonable.
PLN again bases its objection to the book's exclusion on a comparative analysis with other books that contain depictions of sex with a child or discuss erotic attraction to children. TDCJ does not directly respond to these comparisons, but it is relevant that the specific examples PLN relies upon are fictional works—Lolita and Death in Venice—which might make them less troubling than the biographical account in Women Behind Bars.
Though it seems unlikely that the brief, non-explicit passage in Women Behind Bars would create a significant threat to a prison's legitimate goals, in light of the Supreme Court's recognition in Thornburgh that “prison officials may well conclude that certain proposed interactions, though seemingly innocuous to laymen, have potentially significant implications for ․ order and security,” 490 U.S. at 407, PLN must do more to meet its burden than merely assert that TDCJ's policy is unreasonable. As with the other books, PLN has not produced evidence that contradicts the rationality of TDCJ's exclusion of Women Behind Bars.
iv. Lockdown America
TDCJ denied this title because it contains racial slurs and descriptions of racial tensions in prisons, including racially motivated crime by guards against prisoners. Smith testified that books describing racial tensions present a threat of violence because of the existence of race-based prison gangs and the prevalence of racial discord. The specific passage cited to justify the exclusion decision reads: “Another gang of screws [prison guards] at the California Institute for Men at Chino called itself SPONGE, a disgusting acronym for the equally disgusting name, ‘Society for the Prevention of Niggers Getting Everything.’ ” A one-letter notation on the book's database page indicates that it was denied because it “contains material that a reasonable person would construe as written solely for the purpose of communicating information designed to achieve the breakdown of prisons through offender disruption.”5
PLN essentially argues that TDCJ's purported concern over exacerbating racial tensions is merely a pretext for censoring a book that is sharply critical of the prison system. In support of this argument, PLN notes that certain overtly racist books—including Hitler's Mein Kampf and David Duke's Jewish Supremacism—are permitted into TDCJ facilities. In contrast to these obviously racist books, PLN argues that Lockdown America is “anti-racist,” as it criticizes the racism that afflicts many prisons. The distinctive aspect of Lockdown America is that it discusses racial tensions specifically in the prison context.
We recognize the possibility that censorship of viewpoints critical of prison systems or prison administrators may disguise itself in a policy of excluding books depicting guard-prisoner conflict on the purported basis that they are likely to incite similar conflict. But we cannot say that it is unreasonable for prison administrators to conclude that books describing racial tensions in the prison context—as opposed to racial tensions more generally—are more likely to provoke prison violence. Thus, in the absence of specific evidence indicating that TDCJ excluded the book to censor its criticism of prisons, as opposed to excluding it because of its potentially harmful effects, TDCJ's decision does not run afoul of Turner.6 We therefore agree with TDCJ that it is relevant that TDCJ allows a multitude of publications critical of prisons into its facilities, including PLN's newsletter and the majority of the books it sends to prisoners.
b. Other Turner Factors
The remaining Turner factors decidedly favor TDCJ.
i. Alternative means of exercising the right
The second Turner factor—“whether there are alternative means of exercising the right that remain open”—weighs heavily in favor of TDCJ. The challenged policy leaves prisoners and PLN ample alternatives for exercising their First Amendment speech rights. In evaluating the availability of alternatives, the Supreme Court has emphasized that the right in question must be construed “expansively,” meaning that adequate alternatives need not be perfect substitutes for the curtailed right. Thus, in Turner, this factor counted in the government's favor where the regulation forbidding all inmate-to-inmate correspondence did “not deprive prisoners of all means of expression. Rather, it bar[red] communication only with a limited class of other people․” 482 U.S. at 92 (emphasis added). Similarly, in Thornburgh the Court held that this factor was “clearly satisfied” because the challenged regulation, which gave federal prison wardens discretion to censor publications for content-based reasons, “permit[ted] a broad range of publications to be sent, received, and read.” 490 U.S. at 418. And in O'Lone v. Estate of Shabazz, 482 U.S. 342, 352, 107 S.Ct. 2400, 96 L.Ed.2d 282 (1987), in evaluating a regulation that prevented Muslim inmates from attending Jumu‘ah, a religious ceremony necessarily held at a specific time on Friday afternoons, the Court stated that the relevant question was “whether under these regulations respondents retain[ed] the ability to participate in other Muslim religious ceremonies.” (emphasis added).
TDCJ's policy leaves open ample alternative avenues for PLN to express its views to Texas inmates. Prisoners are permitted to read PLN's newsletter and the majority of the books it distributes. Moreover, PLN is free to distribute countless other books that are allowed into the prisons and may choose to distribute new books that it considers suitable substitutes for communicating its message. PLN has not demonstrated that anything unique about the five books in question is fundamental to the communication of its message, and even if it had, that would not be controlling: the Court found adequate alternatives even in Turner and O'Lone, where prisoners were cut off from unique and irreplaceable activities—communicating with specific individuals and participating in a unique religious ceremony. See generally Thornburgh, 490 U.S. at 417–18. The alternatives left open to PLN to communicate its intended message to TDCJ inmates are extensive, providing countless options that must be considered more adequate substitutes for the excluded books than the substitutes considered sufficient in Turner and O'Lone. This factor therefore strongly indicates that TDCJ's practices are reasonable.
ii. Impact on others
“The third factor to be addressed under the Turner analysis is the impact that accommodation of the asserted constitutional right will have on others (guards and inmates) in the prison.” Id. at 418. This factor sheds minimal additional light on the general reasonableness analysis. As the Court explained in Thornburgh: Where “the class of publications to be excluded is limited to those found potentially detrimental to order and security ․ [and][w]here ․ the right in question ‘can be exercised only at the cost of significantly less liberty and safety for everyone else, guards and other prisoners alike,’ the courts should defer to the ‘informed discretion of corrections officials.’ ” Id. (quoting Turner, 482 U.S. at 90, 92) (internal citations omitted). In essence, because TDCJ's regulations are reasonably connected to order and security, this factor is satisfied.
iii. Easy alternatives
[T]he existence of obvious, easy alternatives may be evidence that the regulation is not reasonable․ [I]f a[ ] [plaintiff] 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.
Turner, 482 U.S. at 90–91. PLN has not even suggested an alternative way of accomplishing the same penological objectives promoted by the challenged policy; rather, PLN has only suggested that little harm would result if TDCJ interpreted its policy more narrowly and simply allowed prisoners to receive the contested books. On this point, PLN's argument merely rehashes its disagreement with TDCJ's assessment of the threat posed by allowing the books into the prisons. Because PLN has not proposed an “obvious, easy” alternative, and none is apparent, this factor also weighs in favor of TDCJ.
c. Other Prisons' Policies
The Supreme Court has stated that “the policies followed at other well-run institutions [are] relevant to a determination of the need for a particular type of restriction.” Martinez, 416 U.S. at 414 n. 14.7 PLN argues that with the exception of Soledad Brother, no other prison system in the country censors the books at issue, but this assertion is not fully supported by evidence in the record. PLN relies only on its own experience and on the personal knowledge of three of its authors to substantiate its claim. But as PLN is at pains to point out elsewhere in its brief, it could very well be unaware that certain prisons are censoring its books.8 PLN has not presented enough evidence on the practices of other prison systems for this consideration to influence the resolution of its claims.
d. Consistency and Neutrality
PLN argues that TDCJ's policy, taken as a whole, is not reasonably related to legitimate objectives because it is applied inconsistently. TDCJ essentially concedes that the policy generates inconsistent results, but attempts to argue that inconsistencies are irrelevant to the Turner reasonableness analysis. Both parties are incorrect.
The inconsistent results of TDCJ's policy are partly attributable to the enormous diffusion of responsibility for making initial censorship decisions. The roughly 500 mailroom employees at TDCJ's 96 prison units are all responsible for making those initial decisions, and in the case of approvals, there is no procedure for ever reviewing their decisions. Thus, as TDCJ admits, some inconsistencies are inevitable. TDCJ defends this outcome by arguing that the price of greater consistency would likely be more censorship.
TDCJ relies on Thornburgh for the proposition that inconsistencies are not grounds for granting PLN relief: “The exercise of discretion called for by these regulations may produce seeming ‘inconsistencies,’ but what may appear to be inconsistent results are not necessarily signs of arbitrariness or irrationality.” 490 U.S. at 417 n. 15. But the regulations considered in Thornburgh were crucially different from those considered here. In Thornburgh, the Federal Bureau of Prisons (“FBP”) regulations gave individual wardens the discretion to exclude books for content-based reasons from their own facilities. In that context, certain inconsistencies arose across the various facilities within the FBP. The Supreme Court concluded that such inconsistencies were not fatal precisely because the regulations allowed for facility-specific analysis that accounted for the particular circumstances in which each warden was situated. Id. at 417 (noting that the FBP wardens made their censorship decisions “under the conditions of a particular prison at a particular time”). In this regard, the TDCJ regulations are completely unlike those addressed in Thornburgh: rather than allowing a policy-maker to exercise discretion using a context-specific analysis applicable only to his own facility, TDCJ's censorship decisions apply indefinitely across all 96 of its units. Thus, any inconsistencies resulting from TDCJ's procedures are not subject to the same type of explanation as were the “seeming inconsistencies” discussed in Thornburgh. Id. at 417 n. 15 (emphasis added). Moreover, the Court, which considered only the facial validity of the challenged FBP regulations, expressly noted that whether “variability in enforcement of the regulations stems solely from the censors' subjective views” was relevant to the as-applied challenge that would “be considered on remand.” Id.
On the other hand, the “seeming inconsistencies” addressed in Thornburgh were, at least in one sense, more obviously inconsistent than those about which PLN complains: in Thornburgh, certain federal prisons had excluded the very same book that others had allowed. This type of inconsistency does not exist in the TDCJ system because its exclusion decisions apply system-wide. Thus, the alleged inconsistencies at issue here are only arguable. Concluding that it is inconsistent to allow The Shawshank Redemption while excluding Prison Masculinities requires a subjective assessment of each book rather than a simple matching of the books' ISBN codes. These are precisely the types of subjective assessments that usually fall within prison administrators' discretion.
Ultimately, a limited amount of inconsistency at the margins of TDCJ's exclusion decisions is not enough to defeat the reasonableness of TDCJ's practices. PLN's individualized challenges must be viewed in the context of TDCJ's broader book-review scheme, which certainly bears a reasonable relationship to valid penological objectives. We recognize the possibility that inconsistencies could become so significant that they amount to a practical randomness that destroys the relationship between a regulation and its legitimate penological objectives, but PLN has not presented anything close to enough evidence to demonstrate such a severe problem with TDCJ's policies. We decline PLN's invitation to engage in one-to-one comparisons of a few specific books. While PLN is likely correct that TDCJ's procedures and practices could be improved, we agree with TDCJ's contention that the Constitution does not create a bright line concerning which specific books must be admitted into TDCJ's prisons. Numerous books will fall into a range of discretion, and it follows that TDCJ may reasonably exclude some books and allow others that may be judged to fall in that discretionary range.
PLN has not presented sufficient evidence to survive summary judgment on its First Amendment claims. PLN has, at most, demonstrated that reasonable minds might differ on whether to permit certain books into a general prison population, which is very different from demonstrating that TDCJ's practices and exclusion decisions bear no reasonable relation to valid penological objectives. The principal Supreme Court precedents applicable to PLN's First Amendment claims, which all dealt with facial challenges to prison regulations, emphasize the need for according deference to the judgment of prison administrators, and we conclude that such deference must be at its zenith in the context of challenges to individualized decisions implementing a facially constitutional policy. Any other conclusion would require the federal courts to sit as permanent appeals councils reviewing every individual censorship decision made by state corrections institutions.
II. Due Process Claims
PLN argues TDCJ's book review procedures violate its due process rights by failing to provide notice of denials occurring after a final denial by MSCP/DRC. The district court dismissed PLN's due process claims as moot based on a recent change in TDCJ's written policy concerning its review procedures. As explained above, in 2010 TDCJ amended the Policy to provide that denial notices should be sent to the “sender” rather than “editor and/or publisher,” an amendment that TDCJ asserts brought its written policy into line with its longstanding practice. The district court concluded that this change in policy, by providing notice and the right to appeal to senders, mooted PLN's prospective due process claims for injunctive relief.
PLN correctly argues that the district court misapprehended the significance of the change to TDCJ's notice procedures. The district court apparently understood the change in policy to mean that a denial notice, and the accompanying right to appeal, would be sent to each sender of a book. TDCJ concedes that this is not the case, as it continues its existing practice of sending denial notices only until a final denial decision has been made by the MSCP/DRC and “does not permit further appeals about [a DRC-denied] book.”
PLN argues that because TDCJ's censorship program curtails a liberty interest protected by the First Amendment, it is entitled to due process protections before TDCJ denies a book that it distributes. PLN specifically contends that it has an individualized right to notice and appeal the denial of a book it distributes, regardless of whether other senders have already appealed the same book's denial. TDCJ responds that requiring repetitive appeals for every new sender of a denied book would be burdensome and pointless, considering that the book review process is not individualized to the specific sender or intended recipient, but rather focuses only on whether the book's contents are suitable for the prison population generally.
PLN cites numerous cases recognizing that due process protections apply to certain censorship decisions in the prison context. In Procunier v. Martinez, the Supreme Court held that senders and addressees were entitled to basic due process protections—notice and an opportunity to be heard—before the California Department of Corrections could censor personal correspondence. 416 U.S. at 417–19. Decisions of the Fourth, Ninth, Tenth, and Eleventh Circuits have cited Martinez in recognizing that due process must accompany various decisions to exclude prison mailings. None of these cases, however, addressed the precise question at issue—whether subsequent senders of identical publications are entitled to notice and an opportunity to appeal a denial.
Martinez itself is readily distinguishable. In that case, the Supreme Court considered whether procedural protections were required before a prison could censor personal letters between prisoners and outsiders. In such a case, each censorship decision was necessarily individualized: for each letter considered, a unique screening determination had to be made. No two letters were exactly alike, so the specifics of each one had to be evaluated before a censorship decision could be justified. On the other hand, TDCJ's contested policies pertain to censorship of mass-market books that are applicable throughout the Texas prison system, and censorship decisions are not individualized. Once a book is deemed inappropriate for the prison system, there is nothing new to evaluate the next time a copy of the book is sent to an inmate. The procedural requirements proscribed in Martinez are not therefore necessarily applicable.
PLN cites cases from other circuits that extend Martinez to other types of mail beyond personal letters or other individualized correspondence. In Montcalm Publishing Corp. v. Beck, 80 F.3d 105 (4th Cir.1996), the Fourth Circuit, relying on Martinez, extended due process protections to publishers whose magazines were excluded by the Virginia Department of Corrections (“VDOC”). Although the Fourth Circuit recognized that Martinez “specifically limited its consideration to regulations of ‘direct personal correspondence between inmates and those who have a particularized interest in communicating with them,’ ” id. at 107 (quoting Martinez, 416 U.S. at 408), it stated: “Having recognized that [the magazine publisher] indeed has a constitutional interest in communicating with its inmate-subscribers, we cannot conclude that it is entitled to no process at all when VDOC prevents the subscribers from receiving Montcalm's publications.” Id. at 109.
In Jacklovich v. Simmons, the Tenth Circuit, in considering non-content based regulations limiting prisoner access to publications, agreed that “publishers are entitled to notice and an opportunity to be heard when their publications are disapproved for receipt by inmate-subscribers.” 392 F.3d 420, 433 (10th Cir.2004) (quoting Montcalm Publ'g, 80 F.3d at 106). The Ninth Circuit held that PLN was entitled to due process protections before its newsletter could be banned because it was considered bulk mail due to its postage rate. See Prison Legal News v. Cook, 238 F.3d 1145, 1152–53 (9th Cir.2001).
TDCJ is correct that none of these cases directly addresses the question at issue—whether subsequent senders are entitled to notice and appeal even after an identical publication has been finally denied. TDCJ also is correct that the proper standard of review, contrary to PLN's suggestion, is the standard articulated in Turner, discussed above. See Washington v. Harper, 494 U.S. 210, 224, 110 S.Ct. 1028, 108 L.Ed.2d 178 (1990) (applying the Turner standard to due process claims and emphasizing that it is “quite clear that the standard of review we adopted in Turner applies to all circumstances in which the needs of prison administration implicate constitutional rights”). Thus, TDCJ's procedures need not fully satisfy the traditional demands of due process if the procedures it employs are reasonably related to legitimate penological objectives.
TDCJ's procedures are reasonable and do not run afoul of the Due Process Clause. After a DRC-level denial has become final, subsequent denials of identical publications amount to the routine enforcement of a rule with general applicability. Such subsequent denials are non-individualized—they neither reconsider the content of the denied book nor depend on the particular sender or addressee—so it is not even clear that due process is implicated by such decisions. TDCJ must be permitted to pass rules of general application, even ones that limit prisoner rights, without subjecting such rules to repetitive challenges every time they are applied. “There must be a limit to individual argument in such matters if government is to go on.” Bi–Metallic Inv. Co. v. State Bd. of Equalization, 239 U.S. 441, 445, 36 S.Ct. 141, 60 L.Ed. 372 (1915) (“Where a rule of conduct applies to more than a few people, it is impracticable that everyone should have a direct voice in its adoption.”).
If TDCJ's policies have any procedural infirmities, they would relate to the initial DRC-level decision, not the treatment of subsequent senders. As the Fourth, Ninth, and Tenth Circuits have recognized, in the cases cited above, publishers have due process rights when a prison bans their publications. That is because, at the time of the initial denial, an individualized decision (with respect to the book's contents) must be made. At least arguably, TDCJ's DRC-level procedures fall short of the traditional requirements of due process because they do not give the appellant, whether the prisoner or the outsider, the right to participate in DRC's consideration of the appeal, even informally or through written submissions. See Grannis v. Ordean, 234 U.S. 385, 394, 34 S.Ct. 779, 58 L.Ed. 1363 (1914) (“The fundamental requisite of due process of law is the opportunity to be heard.”). MSCP/DRC reviews only the denial form and the book itself in reaching its decision. Importantly, PLN does not challenge the adequacy of the DRC review procedure—it only demands that it be afforded its own “appeal.” Thus, the adequacy of the DRC review procedure is not properly before the court.
Finally, PLN argues that even if it is not entitled to appeal a previously denied book, due process requires that it at least be given notice of all denials. This argument is incorrect. Due process pertains to the right to participate in government decision making. The “notice” required by due process is notice of when, where, and how one can be heard before a decision becomes final. See Londoner v. Denver, 210 U.S. 373, 385, 28 S.Ct. 708, 52 L.Ed. 1103 (1908) (“[D]ue process of law requires that ․ [a party] shall have an opportunity to be heard, of which he must have notice․”). The right to receive notice exists only to effectuate the right to be heard, and therefore is inapplicable where a party has no right to participate in the decision-making process. Giving notice to a sender that his communication has been rejected may be a reasonable courtesy, but such notice is not a requirement of due process.
III. Qualified Immunity
Qualified immunity protects public officers from suit if their conduct does not violate any “clearly established statutory or constitutional rights of which a reasonable person would have known.” Harlow v. Fitzgerald, 457 U.S. 800, 818, 102 S.Ct. 2727, 73 L.Ed.2d 396 (1982). A two-step analysis governs whether public officials are entitled to qualified immunity.
First, we must determine whether the facts, either as the plaintiff alleges or as proved without dispute, establish that the officer violated a clearly established constitutional right․ [I]f the plaintiff has alleged a constitutional violation, the court must next determine whether the official's conduct was objectively unreasonable under established law.
Linbrugger v. Abercia, 363 F.3d 537, 540 (5th Cir.2004) (internal quotation marks and citations omitted). As explained above, TDCJ has not violated PLN's due process rights, and the individual defendants are therefore entitled to qualified immunity.
For the reasons given above, the judgment of the district court is AFFIRMED.
1. The Policy originally stated that the prisoner and “the editor and/or the publisher” would receive notice and an opportunity to appeal. TDCJ alleges that in practice, it always gave notice to the book's sender rather than its editor or publisher. In February 2010, TDCJ amended the Policy to reflect this practice, expressly extending notice and appeal rights to a book's “sender.” Prisoners may receive publications only from publishers and booksellers.
2. The court declined to dismiss PLN's claims regarding another unordered book, Women Behind Bars, because some prisoners had expressed interest in it after PLN had sent it to them unsolicited.
3. In fact, Bantam Books took as its starting point that book distributors have standing to challenge censorship schemes. See 372 U.S. at 64 n. 6.
4. PLN also argues that the intended recipients of its books have “willingly sought” PLN's point of view. PLN contends that it only sent books to prisoners who had subscribed to its magazine or otherwise been in contact with PLN.
5. PLN misconstrues this provision by interpreting it to mean that a book is properly subject to exclusion only if a reasonable person would conclude that, in its entirety, the book's sole purpose is to achieve the breakdown of prisons. Rather, the language of the provision allows for exclusion of a book that contains any material written for such a purpose.
6. This is difficult to slice because what would obviously constitute content-based discrimination outside the prison context is undoubtedly permissible within it. Thus, the distinction we draw is less clear than the distinction between content-neutral and content-based discrimination. Even viewpoint discrimination is permissible in the prison context.
7. TDCJ argues that Thornburgh implicitly held that the practices of other prisons are irrelevant to the reasonableness analysis, but that is an unjustified conclusion. Though Thornburgh provides a potential explanation why a censorship decision might be reasonable despite being an outlier, it does not make the practices of other prisons irrelevant to the reasonableness analysis. TDCJ would likely find it highly relevant if many other prisons also censored PLN's books.
8. Relatedly, PLN offered Ron McAndrew as an expert witness based on his career as a corrections officer and warden in Florida's prison system. McAndrew's report arguably supports PLN's contention that other prisons likely do not find it necessary to censor the books at issue. TDCJ moved to exclude McAndrew's testimony, raising questions about his competency and methodology, but the district court never ruled on its admissibility, making it an unsound basis for reversing summary judgment. Even if admitted, McAndrew's report carries little weight because it is little more than a recitation that he would not have excluded PLN's books for the reasons cited by TDCJ. As the district court correctly noted, “[m]ere disagreements do not demonstrate irrationality of [TDCJ's] decisions.”
EDITH BROWN CLEMENT, Circuit Judge: