James Pesci is a detainee at the Florida Civil Commitment Center (FCCC), a for-profit facility that houses sex offenders involuntarily committed under Florida's Involuntary Civil Commitment of Sexually Violent Predators Act. Pesci is not a prisoner; like the other roughly 600 residents of FCCC, he has already served out his prison sentence. Instead, he is involuntarily committed because the State has determined that he is a "sexually violent predator" likely to engage in future "acts of sexual violence if not confined in a secure facility for long-term control, care, and treatment." Fla. Stat. §§ 394.912 (10)(b) ; 394.915.

Pesci devotes his time to investigative reporting, and during his commitment he has published two monthly newsletters-publications that are highly critical of FCCC. Citing rising tensions between residents and staff, the facility director deemed one of Pesci's publications a security threat and issued a policy banning its possession or distribution. Pesci now circulates a successor newsletter, but he is constrained by a second, facility-wide policy that limits the number of pages that each inmate can print in the FCCC computer lab. Pesci filed a civil rights action under 42 U.S.C. § 1983 , claiming that FCCC's policies violate his expressive freedoms under the First and Fourteenth Amendments. Applying this Court's four-part test *1163 for evaluating a civil detainee's constitutional claims, the district court granted summary judgment in favor of FCCC. After a searching review of the record, and with the benefit of oral argument, we too conclude that the two policies at issue do not violate the First Amendment because they are reasonably related to FCCC's legitimate interests in facility security and conserving resources. We therefore affirm.

I.

A.

For many years, Pesci published a monthly newsletter, Duck Soup , which frequently excoriated FCCC's staff, sex offender treatment program, and conditions of confinement. Pesci envisioned Duck Soup as "the uncensored pulse of the compound," dedicated to exposing "corruption at FCCC." He called GEO Group, Inc., the for-profit corporation then in charge of FCCC, a "criminal organization that has a chronic history of cover-ups, medical neglect and psychological abuse." He accused GEO of cost-cutting the residents' nutrition and medical care to increase profits. In one issue, Pesci called the residents of FCCC "coward[s]" for failing to hold "collective protests" and "demonstrations." Pesci also leveled accusations against FCCC staff members by name. He reported that a particular captain had sexually harassed his female subordinates, accused one lieutenant of racism and excessive force against inmates, suggested that another lieutenant liked to watch residents shower, and insinuated that multiple staff members used illegal drugs-to take just a few examples.

Pesci was originally permitted to upload Duck Soup to an online blog. And inmates could also print hard copies in the FCCC computer lab, so long as they complied with FCCC's general policy on the use of the facility's printers. Under that policy, which had been on the books since 2006, each inmate was allowed to print 20 pages in the computer lab every other day-or 40 pages every other day if the inmate supplied his own paper. In April 2009, however, Timothy Budz-then the facility director of FCCC-issued a policy prohibiting residents from printing hard copies of Duck Soup unless they supplied their own paper. The policy was supposed "to limit resident access to Duck Soup" on grounds that the newsletter was creating "tensions" between residents and staff, undermining staff authority, and disrupting treatment.

According to Budz, Duck Soup "became increasingly inflammatory" even after the 2009 policy was enacted. In the June/July 2010 issue, for instance, Pesci accused a nurse named Margaret Ferrell of intercepting, reading, and stealing an inmate's outgoing mail. Shortly after the June/July issue, "multiple residents" angrily confronted Nurse Ferrell, making her fear for her safety. She attributed the hostile interaction to Pesci's reporting. Budz testified that other staff members lodged complaints as well, and that he "was afraid that violence was going to break out in the facility." In November 2010, Budz issued a new policy, which declared "that Duck Soup was now contraband and prohibited its distribution or possession." Budz testified that after "the banning of Duck Soup, the tensions and hostility around FCCC ... decreased."

Some years later, a changing of the guard took place at FCCC. Dr. Donald Sawyer replaced Budz as facility director, and Correct Care Solutions replaced GEO as the contractor for the facility. After Sawyer took over, Pesci started a successor publication to Duck Soup called The Instigator . According to Pesci, The Instigator 's stated mission is to "bring interesting news to the FCCC population and their families," to "edify the community of what life in FCCC is really like," and to *1164 "advocate the elimination" of sex offender commitment centers. True to that mission, The Instigator has featured articles on Supreme Court cases relating to inmates' rights, encouraged residents to read Florida Law Weekly and join a legal discussion group, and interviewed residents about their faith and perspectives. Sawyer concedes that The Instigator is "less inflammatory" than Duck Soup and that this "toned down newsletter has raised relatively few security concerns."

Pesci is allowed to write, print, and copy The Instigator , but he cannot distribute it as freely as he did Duck Soup. Under Sawyer's leadership, FCCC residents face stricter computer policies. They can no longer access the internet, and they are not allowed to save files-any files-to library computers for other residents to read or print. FCCC also continues to enforce the 2006 page-limit policy, under which each resident can print only 20 pages every other day using FCCC paper. Accordingly, The Instigator -and, to be fair, every other inmate publication except those distributed by staff-sponsored social clubs-is subject to a page limit. The funding for the paper and ink does not come out of FCCC's pocket; it is paid for by a Resident Welfare Fund, which holds money donated to the residents as well as proceeds from the resident commissary. FCCC maintains, and Pesci does not dispute, that the policy is "applied uniformly" to every individual resident.

B.

In July 2010-a few months before Duck Soup was banned-Pesci filed a pro se § 1983 complaint against Budz alleging that the 2009 printing restrictions on Duck Soup violated his First and Fourteenth Amendment rights. The district court concluded that the 2009 policy did not violate Pesci's constitutional rights and granted summary judgment in favor of Budz. The district court did not have occasion to address 2010's all-out ban on Duck Soup . On appeal, this Court clarified the legal standard that should apply to a civil detainee's constitutional claims-a variant of the test articulated by the Supreme Court for evaluating a prisoner's constitutional claims, modified to reflect the non-punitive nature of civil detention-and remanded for the district court to develop the record as to the 2010 ban and evaluate both policies under the appropriate standard. See Pesci v. Budz (Pesci I) , 730 F.3d 1291 , 1295-97 (11th Cir. 2013). On remand, Pesci (this time represented by appointed counsel) initially filed an amended complaint providing more detail on the 2010 Duck Soup ban. He later requested and was granted leave to supplement his complaint with new claims relating to The Instigator as well.

In June 2015, Pesci filed his second amended complaint-the operative complaint in this appeal-against Budz, Sawyer, and various other FCCC affiliates. The second amended complaint brought First and Fourteenth Amendment challenges against three policies: the 2009 printing restrictions on Duck Soup , the subsequent 2010 ban on Duck Soup , and the 2006 page-limit policy as enforced against The Instigator. 1 Applying the legal standard set out by this Court in Pesci's first appeal, the district court determined that all three policies were constitutional and once again granted summary judgment in favor of the FCCC defendants.

*1165 Pesci has appealed. At oral argument, the parties agreed that the 2010 total ban on Duck Soup mooted Pesci's claim regarding the less-stringent 2009 printing restrictions on Duck Soup. As a result, we consider only the constitutionality of the 2010 Duck Soup ban and of the 2006 page-limit policy.

II.

We review the grant of summary judgment de novo, viewing all evidence and drawing all reasonable inferences in favor of the nonmoving party. Ave. CLO Fund, Ltd. v. Bank of Am., N.A., 723 F.3d 1287 , 1293-94 (11th Cir. 2013). Summary judgment is appropriate "if the movant shows that 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).

"By its very terms, this standard provides that the mere existence of some alleged factual dispute between the parties will not defeat an otherwise properly supported motion for summary judgment." Anderson v. Liberty Lobby, Inc ., 477 U.S. 242 , 247-48, 106 S.Ct. 2505 , 91 L.Ed.2d 202 (1986). "Only disputes over facts that might affect the outcome of the suit under the governing law will properly preclude the entry of summary judgment." Id. at 248 , 106 S.Ct. 2505 .

III.

A.

The first time we considered an appeal in Pesci's case, we decided that the appropriate standard against which to measure a civil detainee's constitutional claims was a variant of the standard established by the Supreme Court in Turner v. Safley for reviewing the constitutional claims of prisoners. 482 U.S. 78 , 89-90, 107 S.Ct. 2254 , 96 L.Ed.2d 64 (1987). Today we apply, but do not revisit, that standard.

As the Supreme Court established in Turner , prison walls "do not form a barrier separating prison inmates from the protections of the Constitution." Id. at 84 , 107 S.Ct. 2254 . This Court has also said that a "prisoner does not surrender his constitutional rights at the prison gates," United States v. Mills , 704 F.2d 1553 , 1560 (11th Cir. 1983), and that every "inmate retains those First Amendment rights that are not inconsistent with his status as a prisoner or with the legitimate penological objectives of the corrections system," Lawson v. Singletary , 85 F.3d 502 , 509 (11th Cir. 1996) (per curiam) (quoting Pell v. Procunier , 417 U.S. 817 , 822, 94 S.Ct. 2800 , 41 L.Ed.2d 495 (1974) ). At the same time, the Supreme Court has acknowledged that running a prison "is an inordinately difficult undertaking that requires expertise, planning, and the commitment of resources, all of which are peculiarly within the province of the legislative and executive branches of government." Turner , 482 U.S. at 85 , 107 S.Ct. 2254 . The "formidable task of running a prison" falls to those other two branches, and "separation of powers concerns counsel a policy of judicial restraint" and "deference to the appropriate prison authorities." O'Lone v. Estate of Shabazz , 482 U.S. 342 , 353, 107 S.Ct. 2400 , 96 L.Ed.2d 282 (1987) ; Turner , 482 U.S. at 85, 107 S.Ct. 2254 . Principles of federalism bolster that deference when "a state penal system is involved." Turner , 482 U.S. at 85, 107 S.Ct. 2254 .

In an effort to vindicate both "the need to protect constitutional rights" and the need for "judicial restraint regarding prisoner complaints," Turner set out the ground rules for evaluating prisoners' constitutional claims: When a prison regulation or policy "impinges on inmates' constitutional rights, the regulation is valid if it is reasonably related to legitimate penological interests."

*1166 Id. at 85, 89 , 107 S.Ct. 2254 (internal citation and quotation marks omitted). The Court then identified four "factors that are relevant to, and that serve to channel, the reasonableness inquiry." Thornburgh v. Abbott , 490 U.S. 401 , 414, 109 S.Ct. 1874 , 104 L.Ed.2d 459 (1989) (applying Turner ). They are:

(1) whether there is a "valid, rational connection" between the regulation and a legitimate governmental interest put forward to justify it;

(2) whether there are alternative means of exercising the asserted constitutional right that remain open to the inmates;

(3) whether and the extent to which accommodation of the asserted right will have an impact on prison staff, inmates, and the allocation of prison resources generally; and

(4) whether the regulation represents an "exaggerated response" to prison concerns.

Pope v. Hightower , 101 F.3d 1382 , 1384 (11th Cir. 1996) (quoting Turner , 482 U.S. at 89-91, 107 S.Ct. 2254 ). This reasonableness inquiry recognizes that courts do not sit as super-wardens, and ensures that prison officials, rather than judges, will "make the difficult judgments concerning institutional operations." Turner , 482 U.S. at 89, 107 S.Ct. 2254 (quoting Jones v. North Carolina Prisoners' Labor Union, Inc. , 433 U.S. 119 , 128, 97 S.Ct. 2532 , 53 L.Ed.2d 629 (1977) ). The distribution of burdens between the parties also reflects the difficulties of prison management from the bench: the burden "is not on the State to prove the validity of prison regulations but on the prisoner to disprove it." Overton v. Bazzetta , 539 U.S. 126 , 132, 123 S.Ct. 2162 , 156 L.Ed.2d 162 (2003) ; see also Jones , 433 U.S. at 128 , 97 S.Ct. 2532 (stating that "the burden was not on [the prison] to show affirmatively" that the creation of an inmate union "would constitute a present danger to security and order" (internal citation and quotation marks omitted)).

Many of these same considerations apply in civil commitment scenarios, where courts must also protect constitutional rights while showing appropriate deference to facility administrators. But unlike a prison sentence, civil commitment is purely rehabilitative; it is not a form of punishment. So we determined that "a similar balance should be struck in scrutinizing the constitutional claims of civil detainees," with the standard "modified to reflect the salient differences between civil detention and criminal incarceration." Pesci I , 730 F.3d at 1297 . "Pesci is not a prisoner and the Florida Civil Commitment Center is not a prison." Id. Accordingly, "the range of legitimate governmental interests is narrower here than it is in a prison context." Id. While, for example, "retribution and general deterrence" are "plainly legitimate justifications for prison regulations," they "decidedly are not a proper foundation for the restriction of civil detainees' constitutional rights." Id. ; see Kansas v. Hendricks , 521 U.S. 346 , 369, 117 S.Ct. 2072 , 138 L.Ed.2d 501 (1997) (upholding the constitutionality of the involuntary confinement of sex offenders in large part because civil detention is "not punitive"). "[I]nstitutional order, safety, and security" remain paramount in the civil detention context, as do "the rehabilitation and treatment of civil detainees." Pesci I , 730 F.3d at 1298 .

Apart from the narrowed universe of justifications, Turner and its progeny govern Pesci's case. Although persons "who have been involuntarily committed are entitled to more considerate treatment and conditions of confinement than criminals," Youngberg v. Romeo , 457 U.S. 307 , 321-22, 102 S.Ct. 2452 , 73 L.Ed.2d 28 (1982), "this observation does not warrant" further "departure from the Turner standard,"

*1167 Pesci I , 730 F.3d at 1298 . "At the same time, deference to the professional judgment of the facility administration is not tantamount to carte blanche permission to deny the fundamental rights of free speech and free expression." Id. at 1299 . The " Turner standard is a deferential one, but it is not toothless." Id. "Deference to facility administrators and concerns relating to safety and security cannot be used as a pretext to silence undesirable speech." Id. at 1300 .

With these directives in mind, we now consider whether there is a genuine dispute of material fact regarding the constitutionality of the ban on Duck Soup and the page limit restrictions on The Instigator . Because the two policies operate somewhat differently, we apply the four-part Turner test to each in turn.

B.

We begin with the 2010 policy that banned Duck Soup. Under the modified Turner standard, the over-arching inquiry is whether the ban on Duck Soup is "reasonably related" to a "substantial governmental interest unrelated to the suppression of expression." Thornburgh , 490 U.S. at 413, 415 , 109 S.Ct. 1874 (internal quotation marks and citations omitted). Timothy Budz, the FCCC administrator responsible for the ban, testified that he believed Duck Soup to be a security threat. He explained that as Duck Soup denigrated individual staff members by name in "increasingly inflammatory" stories, he became "afraid that violence was going to break out in the facility." 2 Pesci does not dispute that maintaining security is a legitimate government objective, and indeed is central to FCCC's mandate under Florida law. See Fla. Stat. § 394.917 (2) (requiring that sexually violent predators be housed in a "secure facility" for "control, care, treatment, and rehabilitation"). So the question to be determined by the four Turner factors is whether the ban on Duck Soup is "reasonably related" to FCCC's legitimate security interest.

1.

"First and foremost," we must determine whether there is a "rational connection" between FCCC's decision to ban Duck Soup and its stated goal of ensuring security in the facility for inmates and staff. Shaw v. Murphy , 532 U.S. 223 , 229, 121 S.Ct. 1475 , 149 L.Ed.2d 420 (2001) (quoting Turner , 482 U.S. at 89, 107 S.Ct. 2254 ). Mindful that we do not sit "as a super-warden to second-guess the decisions of the real wardens," we still must be sure that "more than a formalistic logical connection" exists between the policy and the problems it purports to solve. Prison Legal News v. Sec'y, Fla. Dep't of Corr ., 890 F.3d 954 , 965 (11th Cir. 2018) (quoting Beard v. Banks , 548 U.S. 521 , 535, 126 S.Ct. 2572 , 165 L.Ed.2d 697 (2006) (plurality opinion)). This factor is the most important of the four, and indeed must be satisfied if the policy is to survive. See, e.g. , Beard , 548 U.S. at 532-33 , 126 S.Ct. 2572 ; Crime Justice & Am., Inc. v. Honea , 876 F.3d 966 , 973 (9th Cir. 2017) ; Hatim v. Obama , 760 F.3d 54 , 59 (D.C. Cir. 2014). "If the connection between the regulation and the asserted goal is 'arbitrary or irrational,' then the regulation fails, irrespective of whether the other factors tilt in its favor." Shaw , 532 U.S. at 229-30 , 121 S.Ct. 1475 (quoting Turner , 482 U.S. at 89, 107 S.Ct. 2254 ).

We are persuaded that FCCC's decision to ban Duck Soup was *1168 "rationally connected to its security and safety interests." Prison Legal News , 890 F.3d at 970 . As the Supreme Court recognized in Turner , "prison officials must be able to ' anticipate security problems and ... adopt innovative solutions' to those problems to manage a prison effectively." Id. at 968 (quoting Turner , 482 U.S. at 89, 107 S.Ct. 2254 ). We have squarely rejected the "misconception" that prison officials are "required to adduce specific evidence of a causal link between a prison policy and actual incidents of violence." Id. (internal punctuation omitted) (quoting Lawson , 85 F.3d at 513 n.15 ). "Responsible prison officials must be permitted to take reasonable steps to forestall such a threat, and they must be permitted to act before the time when they can compile a dossier on the eve of a riot." Jones , 433 U.S. at 132-33 , 97 S.Ct. 2532 ; see also, e.g. , Prison Legal News v. Livingston , 683 F.3d 201 , 218 (5th Cir. 2012) (upholding a prison's decision to ban "books describing racial tensions in the prison context" because they were "likely to provoke prison violence"); Singer v. Raemisch , 593 F.3d 529 , 536 (7th Cir. 2010) (explaining that the "question is not whether [the game Dungeons and Dragons] has led to gang behavior in the past" but "whether the prison officials are rational in their belief that, if left unchecked, D & D could lead to gang behavior among inmates and undermine prison security in the future").

Here, Pesci has not met his burden of showing that the ban on Duck Soup was not reasonably related to the FCCC's interest in safety and security. On this narrow question, we see no genuine issue of material fact. To the contrary, unrebutted evidence-including evidence from Pesci-made it rational to think that "limiting inmates' exposure" to inflammatory reports of staff malfeasance could "reduce the risk that inmates will engage in behavior that endangers other inmates, guards," and FCCC staff. Prison Legal News , 890 F.3d at 968 . In fact, Pesci's own expert, Dr. Harry Krop, testified that certain types of stories could pose a security threat. Counsel set out several scenarios-all based on stories that Pesci had previously published in Duck Soup -for Dr. Krop to respond to. Dr. Krop admitted, for example, that a story accusing a "specifically named" lieutenant of racism "could pose ... a security issue for the lieutenant." He also conceded that a story reporting that FCCC staff did not review security footage "could actually encourage crime in the facility." He agreed that a story accusing FCCC of cost-cutting at the residents' expense "could raise the hostility and tension" around the center, and that residents "with violent tendencies" could "act out" after reading such a story. And he acknowledged that a story "claiming that the civil commitment center was responsible for a resident's death" "could raise animosity" between the residents and the staff.

Again, these were precisely the sort of stories that appeared in Duck Soup. Pesci had a track record of publishing incendiary stories about FCCC staff members-accusing them by name of racism, voyeurism, medical negligence, physical abuse, and other bad acts-and he made clear in his own words that he had no intention of stopping. Indeed, in the issue of Duck Soup circulated just before the ban, Pesci promised his readers that they could "look forward to" more investigative reporting in future editions of Duck Soup. Turning his attention to FCCC administrators, he boasted, "I have agents on the compound who are as deeply undercover as some of your sexual preferences and will always wet my beak and be my eyes and ears."

The fact that, in the estimation of even Pesci's own expert, these types of stories could create a safety issue is dispositive. FCCC does not need to prove that a safety *1169 issue had already manifested itself. There was every reason to think that if left unchecked, Duck Soup would continue to publish content that Pesci's own expert agreed could foment "hostility and tension" in a facility full of violent sex offenders-including some with "psychopathic traits" and "impulse disorders." More importantly, at this summary judgment stage, Pesci has not presented any evidence that these risks do not exist-he disputes only that incidents have occurred, not that they could occur. It was rational for FCCC administrators to believe that banning Duck Soup would reduce the possibility of violence between residents and FCCC staff. Again, officials are not consigned to wait for a riot to break out before they can take steps to quell it.

We add that although Budz was not required to "produce evidence of a past incident to satisfy the first Turner factor," he nevertheless did argue that Duck Soup had already caused a hostile encounter between residents and Nurse Margaret Ferrell. Prison Legal News , 890 F.3d at 968 (citing Perry v. Sec'y, Fla. Dep't of Corr. , 664 F.3d 1359 , 1362, 1366 (11th Cir. 2011) ). Nurse Ferrell testified-and Pesci does not dispute-that after the June/July 2010 issue of Duck Soup accused her by name of opening a resident's mail, she was "accosted," "yell[ed] at," and "threaten[ed]" by "multiple residents." She also testified that Duck Soup "put [her] in fear for [her] safety at FCCC."

Pesci does not deny that residents accosted Nurse Ferrell, or that the incident occurred after he published a negative story about her. He instead argues that FCCC "could not conclusively show that this encounter was caused by the Duck Soup story or if the residents were simply angry about the swirling rumors of mail tampering on which Pesci was reporting." He further argues that this "dispute of material fact in the record as to whether Duck Soup played any role in the incident" with Nurse Ferrell should preclude the grant of summary judgment.

To state what should be obvious: An inmate is entitled to the safeguards of Rule 56 just like any other non-moving party, and no amount of deference to prison officials will cause this Court to overlook a genuine dispute of material fact. But in this case, any dispute over Duck Soup 's role in the nurse incident is not material, because the first Turner factor does not require "specific evidence of a causal link between a prison policy and actual incidents of violence." Prison Legal News , 890 F.3d at 968 (internal punctuation omitted) (quoting Lawson , 85 F.3d at 513 n.15 ). "Requiring proof of such a correlation constitutes insufficient deference to the judgment of the prison authorities with respect to security needs." Lawson , 85 F.3d at 513 n.15. Budz testified that his primary concern in enacting the ban was forward-looking ; he stated that Duck Soup "continued to raise animosity, hostility and anger" after the Nurse Ferrell incident, and that he "was afraid that violence was going to break out in the facility." The fact that Duck Soup might have played a part in a previous threatening encounter between residents and a staff member is simply additional evidence that the ban was rationally related to FCCC's security interests.

Finally, Pesci insinuates that the ban on Duck Soup was not truly motivated by concern for facility security. If there was any evidence that FCCC administrators invoked security concerns as a mere "pretext to silence undesirable speech," that would certainly give us pause. Pesci I , 730 F.3d at 1300 ; see also Baraldini v. Thornburgh , 884 F.2d 615 , 620 (D.C. Cir. 1989) (emphasizing that a "reviewing court must always be careful to make certain that prison administrators are not pretextually *1170 using alleged concerns in order to punish an inmate for his or her political or other views"). Here, however, Pesci has presented no evidence of pretext. Instead, he would have us infer that because Budz did not ban Duck Soup until nearly four months after the incident with Nurse Ferrell, Budz was not "truly concerned with security."

But Pesci's fixation with the Nurse Ferrell incident is again misplaced, because Budz never claimed to have banned Duck Soup in direct response to that incident. Rather, Budz testified-and Pesci does not dispute-that "complaints from staff ... continued to accumulate" in the months after the incident, and, once again, that he was "afraid that violence was going to break out in the facility." We are not required to draw far-fetched or unreasonable inferences in Pesci's favor. Graham v. State Farm Mut. Ins. Co ., 193 F.3d 1274 , 1282 (11th Cir. 1999) (per curiam) ("The nonmovant need not be given the benefit of every inference but only of every reasonable inference."). All of the evidence-as opposed to the rhetoric-from both sides suggests that Budz was genuinely concerned for the safety of his staff, and it would be unreasonable to infer otherwise. Accordingly, the first Turner factor favors FCCC.

2.

We conclude that the other three factors weigh in favor of FCCC as well.

The second Turner factor asks whether, in spite of the ban on Duck Soup , Pesci has "alternative means of exercising" his asserted right. Turner , 482 U.S. at 90, 107 S.Ct. 2254 . "When considering this factor, the Supreme Court has instructed that the right must be viewed sensibly and expansively." Pope , 101 F.3d at 1385 (citing Thornburgh , 490 U.S. at 417 , 109 S.Ct. 1874 ). The Court has "found adequate alternatives" even where prisoners are "cut off from unique and irreplaceable activities." Livingston , 683 F.3d at 219 ; see O'Lone , 482 U.S. at 352, 107 S.Ct. 2400 (holding that a policy satisfied this factor even though inmates could not attend a particular Muslim prayer service because the inmates could still "participate in other Muslim religious ceremonies"); Turner , 482 U.S. at 92, 107 S.Ct. 2254 (concluding as to this factor that a regulation prohibiting communication between inmates at different prisons did "not deprive prisoners of all means of expression").

Viewed sensibly and expansively, the right at issue here is Pesci's First Amendment right to communicate his views about civil detention and the conditions of confinement at FCCC-and based on the record, it is clear to us that Pesci indeed has "other avenues" of exercising that right. Turner , 482 U.S. at 90, 107 S.Ct. 2254 (quoting Jones , 433 U.S. at 131 , 97 S.Ct. 2532 ). To begin, he does not claim that FCCC has otherwise restricted his ability to communicate his ideas and allegations. And since at least 2013, FCCC has permitted Pesci to publish The Instigator , a successor publication to Duck Soup. Although Pesci employs less inflammatory rhetoric in The Instigator than he did in Duck Soup , the new publication still serves as a platform for Pesci to lodge grievances against FCCC and express his political views on civil detention more generally. To the extent that Pesci has had to temper his tone in The Instigator , no FCCC policy prevents Pesci from verbally communicating his views to his fellow residents as forcefully as he chooses. Of course, communicating through "alternate publications might not be 'ideal' for [Pesci], but Turner does not demand the ideal." Prison Legal News , 890 F.3d at 973 (noting that while Prison Legal News was barred from distributing its monthly magazine to inmates, it could still send the inmates "a variety of *1171 books"); see also Giano v. Senkowski , 54 F.3d 1050 , 1056 (2d Cir. 1995) (stating that "romantic letters" could be an adequate alternative to "semi-nude personal photographs" of inmates' wives and girlfriends, which were banned). So the second factor also indicates that the ban on Duck Soup is reasonable.

As for the third Turner factor, which requires us to consider the impact that accommodation of the asserted right would have on the facility, FCCC enacted the Duck Soup ban precisely because it believed that accommodating Duck Soup would have a negative impact on guards, staff, and other inmates. Cf. Beard , 548 U.S. at 532 , 126 S.Ct. 2572 ("That circumstance is also inherent in the nature of the Policy: If the Policy (in the authorities' view) helps to produce better behavior, then its absence (in the authorities' view) will help to produce worse behavior ...."). Specifically, FCCC administrators testified that allowing residents to read Duck Soup could increase tension and hostility, potentially resulting in inmate-on-staff violence. The third factor favors FCCC.

The fourth and final Turner factor requires us to decide whether the ban on Duck Soup is an "exaggerated response" to FCCC's security concerns. Turner , 482 U.S. at 90, 107 S.Ct. 2254 . The "existence of obvious, easy alternatives may be evidence that the regulation is not reasonable." Id. But at the same time, Turner does not impose a least restrictive means test: "prison officials do not have to set up and then shoot down every conceivable alternative method of accommodating the claimant's constitutional complaint." Id. at 90-91 , 107 S.Ct. 2254 .

Pesci argues that FCCC's security concerns "could likely have been resolved by asking Pesci to refrain from engaging in certain discussions that might impede treatment or security." He points out that Dr. Sawyer, who replaced Budz as the director of FCCC, once sent Pesci a letter requesting that he not discuss "residents by name" or give "details of their personal and private medical needs and care" in The Instigator -and suggests that Budz could have taken a similar approach instead of banning Duck Soup outright .

We decline to hold, and indeed find it inconceivable, that FCCC was required to address a security threat by making polite requests. Cf. Prison Legal News , 890 F.3d at 975 ("Why not simply post signs reminding inmates not to escape?"). We also note that Budz initially did attempt to regulate Duck Soup through a less restrictive printing policy, which failed to achieve the desired results. Budz only banned Duck Soup after concluding that the printing restriction had failed to "quell the rising tensions and difficulties encumbering FCCC." Whether or not we would choose a Duck Soup ban as the best tactic for reducing, as Budz put it, "animosity, hostility and anger" is irrelevant-we are not "the primary arbiters of what constitutes the best solution to every administrative problem." Turner , 482 U.S. at 89 , 107 S.Ct. 2254 . And "when prison officials are able to demonstrate that they have rejected a less restrictive alternative because of reasonably founded fears that it will lead to greater harm, they succeed in demonstrating that the alternative they in fact selected was not an 'exaggerated response' under Turner ." Thornburgh , 490 U.S. at 419 , 109 S.Ct. 1874 . We therefore cannot say that the ban on Duck Soup was an exaggerated response to FCCC's security concerns.

3.

The 2010 ban on Duck Soup is rationally related to FCCC's legitimate government interest in facility security. The ban does not deny Pesci "all means of expression," and FCCC administrators reasonably believed *1172 that accommodating Pesci's desire to publish inflammatory stories could have led to violence. The ban was not an exaggerated response to security concerns, because a less restrictive regulation had already proved ineffective. Because we see no material factual dispute regarding these factors, and all of the factors favor FCCC, we hold that the 2010 ban on Duck Soup does not violate the First Amendment.

C.

We now apply the same Turner analysis to the 2006 page-limit policy. Notably, it was the FCCC Resident Council-a committee comprised of one elected representative from each resident dormitory-that recommended the 2006 policy, which allows each resident to print no more than 20 pages every other day. The stated goal of the policy is to conserve paper and ink for each of the approximately 630 residents, and to minimize wear and tear on the library printers. The money for the printers, paper, and ink comes out of the Resident Welfare Fund, which is funded by outside donations and sales from the resident commissary. Moreover, residents who supply their own paper may print 40 pages every other day-a fact that supports the conservation rationale behind the policy. Guided by the four Turner factors, we conclude that the 2006 policy is reasonably related to the legitimate goal of conserving facility resources.

First , Pesci does not dispute that a civil commitment center has a legitimate interest in conserving facility resources. See Matherly v. Andrews , 859 F.3d 264 , 283-84 (4th Cir. 2017) (recognizing that "allocating scarce resources in an effective fashion" is a legitimate government interest in a civil detention setting). And we have no trouble concluding that the 2006 page-limit policy is rationally connected to the goal of conserving resources; one obvious clue is that residents who supply their own paper may print additional pages. Another, of course, is that the Resident Council recommended this policy when asked to provide input on how the Resident Welfare Fund ought to be disbursed.

Second , the policy does not deprive the residents of "all means of expression." Turner , 482 U.S. at 92, 107 S.Ct. 2254 . Indeed, the record shows that Pesci still manages to publish The Instigator to a wide audience in spite of the page-limit policy. The March 2014 edition of The Instigator thanked residents for donating stamps and envelopes and "firing out freshly printed copies of the Instigator to sources in the community," and praised residents who "used their monthly allotted free legal postage to send copies of the Instigator" to their "lawyers, judges and State attorneys who need to know what life at the Center is truly like." This factor favors FCCC as well.

Third , as for the potential impact of accommodating Pesci's asserted right, it is common sense that lifting the page-limit policy could negatively affect the facility by depleting the Resident Welfare Fund and wearing out the library printers. We note that even under the current policy, the residents can collectively print over 12,000 pages every other day. We defer to the FCCC staff-who, as it turns out, deferred to the resident's elected representatives-in their assessment that any greater printing privileges would put a strain on scarce resources.

Fourth , the 2006 page-limit policy was not an exaggerated response to concerns about conserving resources. Pesci proposes various alternatives: he argues that FCCC should let him supply his own ink, or better yet, "distribute his newsletter online or on the facility computers." FCCC responds that allowing inmates to bring in personal ink cartridges would be logistically burdensome, and that inmates are not granted internet access or file sharing privileges. It *1173 is not our job to micro-manage the FCCC computer lab. And we are certainly in no position to second-guess the facility's decision to prohibit violent sex offenders from accessing the internet and sharing files.

Pesci also takes issue with the fact that the 2006 page-limit policy does not apply to FCCC's facility-sanctioned "social clubs," which include two "gavel clubs," a "creative arts" club, and a club "devoted to supporting military service." Staff members sponsor these clubs as part of the "therapeutic process" and describe them as "pro-social," to the extent that participation in such clubs builds life skills, friendships, and self-esteem. FCCC permits these social clubs to print collaborative newsletters without a page limit-unlike Pesci, whose personal newsletter, The Instigator , is constrained by printing limitations. Pesci argues that the social club exemption amounts to unconstitutional, content-based discrimination. We disagree.

The First Amendment's neutrality requirement operates differently within prison walls; in many cases, "what would obviously constitute content-based discrimination outside the prison context is undoubtedly permissible within it." Livingston , 683 F.3d at 218 n.6. To take just a few examples, the Supreme Court has upheld federal prison regulations banning any publication that "describes procedures for the construction or use of weapons," "describes methods of escape from correctional facilities, or contains blueprints, drawings or similar descriptions of Bureau of Prisons institutions," or "describes procedures for the brewing of alcoholic beverages, or the manufacture of drugs." Thornburgh , 490 U.S. at 405 n.5, 109 S.Ct. 1874 . The Court explained that even though such "determinations turn, to some extent, on content," they are nevertheless "neutral" in the relevant sense because they "further an important or substantial governmental interest unrelated to the suppression of expression." Id. at 415-16 , 109 S.Ct. 1874 (internal quotation marks and citation omitted).

Pesci argues that the 2006 policy discriminates based on content because "whether an organization may avoid the printing restriction" depends on "whether it meets the nebulous criteria of being 'pro-social' " in the eyes of FCCC officials. But that isn't how the policy is structured. The page-limit policy applies across the board to all individual inmates, and the social club exemption applies across the board to all social clubs. Pesci is ineligible for the exemption not because The Instigator is not "pro-social," but because Pesci is not a social club. Indeed, the 2006 printing limitation is a classic content-neutral policy because it draws no distinctions based on content whatsoever. 3

Because it is content-neutral and satisfies all four Turner factors, the 2006 page-limit policy does not violate Pesci's First Amendment rights.

IV.

Pesci has a First Amendment right to publish his newsletters, but that right is not unlimited. Facility administrators must be able to anticipate security problems, so *1174 we defer to FCCC's decision to ban Duck Soup . And the page-limit policy enforced against The Instigator is clearly related to FCCC's legitimate interest in conserving resources. Because both regulations are valid under the Turner reasonableness standard, we affirm the judgment of the district court.

AFFIRMED.

JORDAN, Circuit Judge, concurring in part and concurring in the judgment.

I join Parts I, II, III.A, III.C, and IV of the majority opinion. As to Part III.B, which addresses the ban on Mr. Pesci's Duck Soup newsletter, I concur in the judgment.

* * * * *

Content-based and viewpoint-based restrictions on speech are generally antithetical to the First Amendment and must therefore satisfy strict scrutiny. See Reed v. Town of Gilbert , --- U.S. ----, 135 S. Ct. 2218 , 2226-27, 192 L.Ed.2d 236 (2015) ; Wollschlaeger v. Governor of Fla. , 848 F.3d 1293 , 1300 (11th Cir. 2017) (en banc). But in a case like this one, involving restrictions imposed in an involuntary commitment setting, strict scrutiny does not apply, and our precedent requires us to use a modified version of the test enunciated in Turner v. Safley , 482 U.S. 78 , 89, 107 S.Ct. 2254 , 96 L.Ed.2d 64 (1987). See Pesci v. Budz , 730 F.3d 1291 , 1297-98 (11th Cir. 2013). Under this standard, and given the summary judgment record before us, I reluctantly agree that the ban on Duck Soup should be upheld.

In the past, we have expressed concerns about blanket bans in the prison setting, and held that administrators "must review the particular issue of the publication in question[.]" Guajardo v. Estelle , 580 F.2d 748 , 762 (5th Cir. 1978). When the Supreme Court later ruled that a prison regulation allowing bans on certain incoming publications is facially constitutional if it is reasonably related to legitimate penological interests, it was "comforted by the individualized nature of the determinations required by the regulation." Thornburgh v. Abbott , 490 U.S. 401 , 416, 109 S.Ct. 1874 , 104 L.Ed.2d 459 (1989). After Abbott , other circuits have emphasized the need for individualized review. See Shakur v. Selsky , 391 F.3d 106 , 115-16 (2d Cir. 2004) ; Kikumura v. Turner , 28 F.3d 592 , 598 (7th Cir. 1994). And we reaffirmed the vitality of Guajardo in Owen v. Wille , 117 F.3d 1235 , 1237-38 (11th Cir. 1997), a post- Abbott case. Suffice it to say that, even in the prison setting, the First Amendment prefers a chisel to a sledgehammer.

As noted, we are not writing on a blank slate. A previous panel announced the standard for First Amendment claims in a case like this one, see Pesci , 730 F.3d at 1297-98 , and we may not deviate from that holding. We are therefore required to defer to "the professional judgment of institutional officials" in the involuntary commitment context. Id. at 1298 . Here there is nothing in the record to directly contest Mr. Budz's professional judgment with respect to the impact of Duck Soup on institutional security, so this deferential approach requires us to uphold the ban on that publication.

* * * * *

At the summary judgment stage, we have correctly not accepted the declarations of Mr. Budz and Dr. Sawyer regarding the purported adverse effects of Duck Soup on treatment at the FCCC. Dr. Krop, Mr. Pesci's expert, opined that Duck Soup did not have a detrimental effect on treatment, and that is sufficient evidence to create an issue of fact. Moreover, a number of residents submitted affidavits rejecting the claims of Dr. Sawyer and Mr. Budz. The treatment rationale simply cannot support the ban on Duck Soup at summary judgment.

*1175 Turning to the security rationale, Mr. Budz said that he "receiv[ed] complaints from staff members and residents regarding what [Mr. Pesci] was writing" and that Dr. Wilson informed him "of the rising tensions between the staff and residents due to the highly inflammatory content in Duck Soup ." D.E. 127-1 at 3. Normally, this would not be enough to justify a wholesale content-based ban that would prevent Mr. Pesci from publishing anything critical like Duck Soup . See Brown v. Phillips , 801 F.3d 849 , 854 (7th Cir. 2015) (explaining that an administrator's bare assertion that it is "common sense" that a ban on sexual material will promote treatment at an involuntary-commitment institution is an insufficient basis on which to grant summary judgment under the modified Turner standard). But it appears that Mr. Budz was never deposed and, as a result, Mr. Pesci was unable to make out the bases for Mr. Budz's assertions or to question their validity. 1

Mr. Pesci, moreover, never sat for a deposition, or filed his own declaration in connection with his opposition to summary judgment on the security rationale. He consequently failed to provide evidence to undercut or contradict Mr. Budz's factual assertions.

Nor did Dr. Krop effectively respond to Mr. Budz's contentions. In his report, Dr. Krop stated the following: "Although I cannot speak to the alleged tension among the facility's employees, the majority of residents interviewed dispute that the newsletters increased tension among the residents and/or between residents and staff." D.E. 131-1 at 1. Given that the FCCC houses approximately 660 residents, see D.E. 127-1 at 2, the limited scope and number of Dr. Krop's interviews did little to challenge Mr. Budz's declaration. And Dr. Krop's concession that Mr. Pesci's writings could inflame tensions further undermined his own testimony (and to an extent supported Mr. Budz's claims). See, e.g., D.E. 139-16 at 32-36.

Finally, as even Mr. Pesci acknowledged, Duck Soup did not always set the benchmark for journalistic standards. See, e.g., D.E. 129-7 at 4 ("Although I am not always right, I've strived to fine tune The Instigator and make it much more accurate than, let's say, the infamous Duck Soup ."); D.E. 129-3 at 1 ("You would think that the administration would take a hands off approach, not only because the publisher has refrained from flirting with inflammatory language or articles that could be misconstrued as inciting mutinous behaviors[.]"); id. at 5 (noting that The Instigator "is more investigative in structure and has developed credible sources amongst both staff and residents alike"). I think it is fair to say, on this record, that Duck Soup included unsubstantiated allegations of misconduct which could cause security problems. The prototypical example is Mr. Pesci insinuating that a lieutenant at the facility enjoyed watching residents in the shower or behind privacy curtains. See D.E. 127-4 at 15. It is not difficult to imagine how these types of accusations could raise tensions at the FCCC or pose a security risk for staff and residents.

* * * * *

Mr. Pesci continues to criticize the FCCC through his subsequent publication, The Instigator . Significantly, Dr. Sawyer's professional judgment is that this "toned down newsletter has raised relatively few security concerns." D.E. 129-1 at 2. Viewed in this context, the ban on Duck Soup is more appropriately viewed as a prohibition *1176 on certain types of statements that, without a basis in fact, accuse the FCCC staff of certain unprofessional conduct or employ a vitriolic tone aimed at stirring up emotions.

A review of what Mr. Pesci has been permitted to publish supports this conclusion. For example, articles in The Instigator have criticized the food service program at the FCCC, see D.E. 129-3 at 4; raised suspicions about "incidents involving both the mail room and package room," see D.E. 129-5 at 3; praised FCCC resident activists working to address "concerns about [Florida's sexually violent predator] law and conditions at the facility," see D.E. 129-4 at 4; criticized proposed legislation regarding sex offenders, see D.E. 129-5 at 4; criticized GEO for failing to provide adequate bonuses to staff members, see D.E. 129-5 at 8; described, in detail, what Mr. Pesci believed was his unjustified confinement in a "stripped room" and subsequent hunger strike, see D.E. 129-7 at 1-3; and referred to a visit from GEO and CCS executives as an "Annual Dog & Pony Show," see D.E. 129-9 at 1. In some ways, then, the ban on Duck Soup is a "ban" in name only.

* * * * *

Had Mr. Budz completely precluded Mr. Pesci from criticizing the FCCC or the laws which keep him there, this would be a different case. But Mr. Pesci has been allowed to publish The Instigator , and on this record the ban on Duck Soup is not unconstitutional under the modified Turner standard.