Krasno, Madeline v. Board of Regents of University of Wisconsin et al

Filing 64

OPINION AND ORDER granting 31 Defendants' Motion for Summary Judgment; denying 34 Plaintiff's Motion for Summary Judgment. Signed by Magistrate Judge Stephen L. Crocker on 11/2/22. (jat)

IN THE UNITED STATES DISTRICT COURT
FOR THE WESTERN DISTRICT OF WISCONSIN

MADELINE KRASNO,
Plaintiff,

OPINION AND ORDER

v.
21-cv-99-slc
JENNIFER MNOOKIN1, CHARLES HOSLET,
JOHN LUCAS, MIKE KLEIN, and
NATE MOLL,
Defendants.
____________________________________________________________________________________
Like most public universities, the University of Wisconsin-Madison uses social media to
communicate with its students, staff, alumni, and the general public. This case concerns the
University’s @uwmadison Instagram and Facebook accounts, sometimes referred to as “pages,”
which are the official and primary Instagram and Facebook accounts for the institution. Plaintiff
Madeline Krasno is an alumna of the University and a self-described advocate for humane
treatment of animals. Krasno sometimes leaves comments on the University’s pages expressing
opposition to the University’s use of primates for research purposes. After discovering that the
University had been deleting her comments or hiding them from public view, Krasno filed this
action under 42 U.S.C. § 1983, contending that University officials and employees were
violating her First Amendment rights to free speech and to petition the government.
Specifically, Krasno alleges that the University engaged in unlawful censorship when: (1) the
University placed an account restriction on her Instagram account from September 2020 to
January 2021; (2) the University manually deleted her December 9, 2020 comment to one of
the University’s Facebook posts; and (3) the University employed (and continues to employ)

1

In accordance with Fed. R. Civ. P. 25(d), the University’s new chancellor, Jennifer Mnookin, is
substituted for her predecessor, Rebecca Blank.

keyword filters on both platforms that she claims prevent or make it difficult for her to post
publically-visible comments that reflect her views on animal testing.
The University admits that it suppressed Krasno’s speech, but it denies that it did so
because of her views on animal testing. Instead, contends the University, it hid Krasno’s
comments because they were “off topic” to the subject of the post on which Krasno was
commenting, a practice it maintains is permissible to maintain the purpose of opening up the
comment threads in the first place. Although the University has removed Krasno’s account
restriction, it defends its right to “hide” or delete off-topic comments, either manually or through
auto-moderation tools offered by Facebook and Instagram.
Both sides have moved for summary judgment. Dkts. 31, 34. I am denying Krasno’s
motion and granting defendants’ motion. As discussed below, because there was no settled law
putting defendants on notice that their moderation decisions violated Krasno’s First Amendment
rights, her individual capacity claims must be dismissed under the doctrine of qualified
immunity. And because the University lifted the account restriction in January 2021, Krasno’s
official capacity claim with respect to that practice is barred by the Eleventh Amendment.
Finally, her claim that she faces harm from the University’s use of the keyword filter is too
speculative and remote to warrant injunctive relief, so I am granting summary judgment to the
University on that claim, as well.
From the parties’ joint and individual proposed findings, I find the following facts to be
undisputed for purposes of deciding the instant motions:

2

FACTS
I. Background
The University of Wisconsin-Madison uses social media as a platform to communicate
with its students, staff, alumni, and the general public. As noted, this case concerns the
University’s @uwmadison Instagram and Facebook accounts, sometimes referred to as “pages,”
which are the official and primary Instagram and Facebook accounts for the institution.
Instagram and Facebook are social media platforms owned by Meta. The platforms
facilitate interactions between users by allowing them to share text, photos and videos, and to
like, share, or comment on other users' shared content. Account holders on both platforms also
can call another user’s attention to a post by “tagging” them, which results in that content
appearing on the other user’s Instagram or Facebook page.
Meta offers its account holders various settings and tools to govern the level of
interaction they wish to allow on their social media webpages. The University has opted to
make its Instagram and Facebook pages “public,” meaning they are accessible to anyone with an
Instagram or Facebook account.2 The University also has opted to turn the “commenting”
feature of the webpages “on,” which allows other Instagram and Facebook users to leave
comments, and to reply to other users’ comments in the interactive area that appears beneath
the University’s posts.
The University uses its Facebook page and Instagram account to communicate official
University announcements, events and policies to its student body and to the public on a wide
range of University-related issues. It sees social media as an opportunity to promote the UW-

2

Members of the public without Instagram or Facebook accounts also are able to access the
University’s social media pages, but they cannot leave comments without creating an account.

3

Madison “brand” and to get information to its key stakeholders quickly and publicly, and as an
opportunity for those stakeholders to access the University. The content included in the
University’s social media posts on its @uwmadison Instagram and Facebook accounts is
represented as the official “voice” of the University.
The subjects of the University’s posts range from feel-good content (e.g., celebrating the
accomplishments of its sports teams and lauding research breakthroughs by its scientists) to
letters from university administrators providing updates and information impacting campus life
(e.g., COVID-19 protection measures, and announcements of university closures).

The

University has posted about the medical care provided by its veterinary and engineering schools,
and has featured animals.
Depending on the topic, posts may generate hundreds or even thousands of responses
in the comment threads, ranging from anodyne emojis and tags to substantive critiques of
University policy. For instance, in response to a December 15, 2020 University Facebook post
describing its mobile COVID-19 testing facility, over 50 comments and replies were made that
debated the necessity, rationale, and efficacy of the University's testing, along with comments
alluding to the University's ability to track the whereabouts of its students. See dkt. 60, exhs.
1-3. The University sometimes engages in its own speech in the comment threads by responding
to questions posed by other users.
Regardless of the number of comments appended to a particular post, the University’s
own posts on both Facebook and Instagram are always visible for engagement with users. In
general, the University does not remove or restrict criticisms from the comment threads
associated with its Facebook posts, even when the criticisms are aimed at University policy or

4

practices on topics such as holding in-person classes, expenditures on guest speakers and on art,
and the University’s solicitation of donations to fund student fees.3
The University’s Office of University Communications is responsible for generating the
content of the University’s posts on its Instagram and Facebook pages, for administrating the
accounts, and for moderating users’ comments. Defendant John Lucas is Assistant Vice
Chancellor for University Communications; Mike Klein is Director for News Content and
Editorial Projects at the University; and Nate Moll is a Social Media Manager. Moll, and to a
lesser extent Klein, are primarily responsible for overseeing operations of the University’s social
media accounts and for moderating comments on its Instagram and Facebook pages. Lucas and
the Office of University Communications are responsible for ensuring that the University’s social
media managers comply with any guidance concerning content moderation.
The University conducts invasive research on animals, including invasive research on
nonhuman primates at both the Harlow Center for Biological Psychology and the Wisconsin
National Primate Research Center. In April 2020, the University was fined for alleged violations
of the Animal Welfare Act, including one involving a traumatic injury to a primate when it was
returned to its shared cage despite a previous act of aggression by its cagemate.
These facts make the University a frequent target of criticism and social media campaigns
by animal rights groups, such as the People for the Ethical Treatment of Animals.

The

University’s Communications team monitors the social media activities of these groups to stay
apprised of what they are saying about the University’s animal testing program, and when a

3

I infer from the parties’ proposed stipulated facts that Krasno does not agree that the University
is similarly tolerant of critical comments posted on Instagram. However, there is no evidence that the
University assesses content differently on Instagram than it does on Facebook. I presume this disagreement
arises from the fact that most of Krasno’s comments were made on Instagram.

5

social media campaign might be afoot. The University Communications team routinely shares
this information with social media managers in the Research Communications department,
whose bailiwick includes communications involving animal research, and they sometimes consult
with Research Communications employees about moderation decisions.

II. The University’s Moderation Policies
At the time this lawsuit was filed, the University’s only written policy concerning what
people could post on its Instagram and Facebook pages was a “Social Media Statement.” This
statement could be found on the University’s website and by clicking on a series of links on its
Instagram and Facebook pages. The policy stated in relevant part:
While UW-Madison does not regularly review content posted to
social media sites, it shall have the right to remove any content for
any reason, including but not limited to content that it deems
threatening, profane, off-topic, commercial or promotion of
organizations or programs not related to or affiliated with the
university, or otherwise injurious or illegal.
After this lawsuit was filed, the University issued Interim Guidance to social media staff.
The Interim Guidance states that the University’s social media managers “may engage in content
moderation of social media pages based on one criterion: whether posted content is on vs.
off topic.” Dkt. 32-2 (emphasis in original). The Interim Guidance further states that social
media managers may remove posts that are “unrelated to the topic or purpose of the page,” and
that to make this determination, the manager is to “evaluat[e] the stated purpose for which the
page exists, and in the context of comments, the subject, topic, or purpose of the initial post to
which the comment is attached.” Id. The guidance further provides that social media managers
cannot moderate content based on the viewpoint it expresses. Id. As an example of permissible
6

“off topic” content moderation versus impermissible “viewpoint” moderation, the Interim
Guidance explains that, “while the social media manager of a page dedicated to UW-Madison's
animal research programs may hide a post stating ‘Taylor Swift is the BOmB,’ he or she cannot
hide a post that states, ‘Taylor Swift agrees that all universities should stop torturing animals by
using them for research. This includes UW-Madison's animal research programs!’” Id.
After the parties submitted their summary judgment materials, the University amended
its Social Media Statement. It now reads, in relevant part:
UW-Madison maintains the right to remove content for [sic] that
is off topic, obscene, a violation of intellectual property rights or
privacy laws, commercial, or otherwise illegal or not germane to the
subject matter of the institutional post.
https://www.wisc.edu/social-media-statement/ (visited Oct. 31, 2022)4.
Moll and the other social media managers regularly monitor the comment threads to see
how their content is generally being received, the reactions a particular post is generating, and
to respond, if warranted, to questions that students may raise about a certain topic. For
example, if the University issues a post about a public health issue or policy and a user has a
question, then the University wants to see that comment. The social media managers also police
the threads for comments that are prohibited by the Social Media Statement, either by hiding
them, or by deleting them if this is deemed appropriate. Comment review by the social media
managers is not always contemporaneous. Moll testified that he typically reviews comments
shortly after the University posts something, then once again later in the day, and then before
bed.

4

I take judicial notice of this fact sua sponte pursuant to F. R. Ev. 201(b)(2) and (c)(1). Although
the change in the Social Media Statement post-dates the parties’ summary judgment submissions, it is
appropriate for the court to consider it because Krasno seeks injunctive relief.

7

Neither the Social Media Statement nor the Interim Guidance requires the removal of
any comments, even if they are found to be “off-topic,” and the University does not expect that
every single comment will be reviewed. Some posts have generated comments numbering in the
hundreds or even thousands, and defendants admit they do not review each and every comment,
nor do they provide 24/7/52 moderation of the Instagram and Facebook pages. Thus, although
Moll and the other social media managers can–and do–remove or hide off-topic comments, some
comments are not reviewed at all.
For those comments they do review, the University’s social media managers have broad
discretion to hide or delete a comment containing content they deem “off-topic.” The decision
to hide or remove a comment as “off-topic” is made by the social media manager in charge of
moderation at that time. According to the University’s Interim Guidance, a social media
manager who is unsure whether a comment is off-topic is to err on the side of leaving it visible,
pending consultation with the University’s Office of Legal Affairs.
In the past, the University has chosen to respond to comments critical of the University
that were made in posts that it determined were unrelated to the comment’s topic. Whether to
hide a comment or respond to it is within the discretion of Moll, Klein and Lucas, who would
make that decision based on the particular circumstances, including the time of day the
comment was read and whether or not a subject matter expert was available to answer questions
about it.

8

III. The Keyword Filter
To assist in moderating its social media pages, the University uses a “keyword filter,”
which is an auto-moderation tool offered by Meta to all of its Facebook and Instagram account
holders. To use the tool, the account holder adds words or phrases to the list that it wants to
“filter out.” If a commenter uses a word or phrase on the keyword filter list, then the comment
containing that word will automatically be hidden from public view unless the University
manually “unhides” it.5 The University’s social media managers add words to the list when they
see a word or phrase consistently being used in off-topic comments that appear to be part of a
spam campaign targeting the University’s social media sites.

The vast majority of these

campaigns relate to animal testing and can generate hundreds of comments. These campaigns
pose a challenge for the University’s social media managers because the large volume of
comments makes manual moderation difficult.
The process of adding or removing words from the keyword filters is situation-dependent.
For example, around June 9, 2021, after more than 200 off-topic comments were made on the
University’s Instagram page that included the phrase, “stop animal test,” Moll added that phrase
to the Instagram keyword filter; he removed it sometime later. In April 2022, PETA Latino ran
a campaign that resulted in thousands of comments on the University’s Instagram page that
included the following phrases: “noexperimentarconanimales,” “Asesinos,” “De los animals,” and
“No más experimentos.” Moll added these words and phrases to the Instagram keyword filter
and later removed them once the comment campaign traffic decreased.

5

Comments hidden through auto-moderation on Facebook remain visible to the commenter’s own
Facebook friends, but this not the case with Instagram.

9

As of April 21, 2022, the University’s Instagram keyword filter list contained the
following words or phrases: #freebaby cocoa, #releasecornelius6, @peta, Cornelius, WNPRC,
abusing, animal testing, biden, cruelty, kill animals, killers, lab, monsters, monstros, rot in hell,
shame on, testing cats, testing on animals, testing on cats, tests on cats, torture, torturing,
trump, vivisection, you guys are sick.7 Dkt. 38-22.
On the University’s Facebook page, the keyword filter list as of April 21, 2022 included
the following words and phrases related to animal testing:
animal laboratories

macaques

torture

animal testing

monkeys

torturing

barbaric

primate

vivisection

cruelty

primates

wnprc

experimenting on

testing on animals

There are 29 other words or phrases on the list, including terms related to law enforcement (such
as “support the police” and “blue lives matter”), profanity, and the words “biden” and “trump.”8
Dkt. 38-22, at 2. The record does not reflect precisely when each word was added to the list.
There is no formal practice or review for removing words from the keyword filters; rather,
words and phrases are reviewed for potential removal on an as-needed basis. If the University
posts content relevant to a word or phrase captured by the filter, for example, then the
University

would remove that word from the filter.

There is no process for reviewing

automatically moderated comments to ensure that they were not wrongly captured by the
6

“Cornelius” is the name of a primate at the Wisconsin National Primate Research Center.

7

The list also contains four profane words.

8

Although the University removed some words and added others after the parties filed their
motions for summary judgment, there appears to be no dispute that the list continues to include words
frequently used by anti-animal testing advocates, or that if it doesn’t, the University might add such words
to the list again in the future.

10

keyword filter. If a social media manager happens to notice a comment that was filtered
erroneously, then he or she can “unhide” that comment if it was made on Facebook, but not on
Instagram.
The University concedes that its moderation system is imperfect. Several examples of offtopic comments are visible in the comment threads to various posts on the University’s social
media pages. See Foley Dec., dkt. 38, exh. 7 (comment by “Lucky Dave”), exh. 11 (comment
by Barbara Luebke Merten) exh. 23 (comment by user “Hani Hosain”), exh. 12 (comments by
users “sara.cheney”and “jatzyy”), exh. 34 (comment by user “lili_varela”), exh. 27 (comment by
user “rollietimothy”), 6-8.

Some of these off-topic comments were unmoderated while

comments relating to animal advocacy were hidden. Dkt. 54, Reply to PPFOF, 37-39. Moll
conceded that the University has hidden some comments relating to animal advocacy even
though they probably were on-topic. Dkt. 54, Reply to PPFOF, ¶¶ 34, 35.
On five occasions since January 1, 2020, the University has turned off its commenting
features on Instagram or Facebook, preventing any user from commenting. Three of these
occasions were due in whole or in part to spam campaigns related to the University’s animal
research program. On two of those three occasions, comments also were disabled because many
portrayed misinformation about COVID-19.

IV. The University’s Moderation of Krasno’s Comments
Plaintiff Madeline Krasno is a former student of the University who worked as a student
primate caretaker at the Harlow Center. Krasno's work at the Harlow Center inspired her to
educate others about what she perceives to be the unknown harms of primate testing, specifically

11

the treatment of primates, and to advocate for the end of animal testing in laboratories. Krasno
does this in part by highlighting her previous experience, and by speaking against animal testing
in her own social media posts. Krasno has never been prevented from sending emails to UWMadison expressing her viewpoint on animal research and in fact, she has communicated such
views via email. Nor has the University restricted Krasno from sending letters, making phone
calls or speaking publicly about her experiences at UW-Madison’s Harlow Lab. However,
Krasno wants to engage with others on UW-Madison’s Instagram and Facebook pages in order
to express her views on animal testing, on funding that the University receives for animal testing,
and on her own experience with animal testing.
On September 18, 2020, the University created a post on its Instagram account about
dairy cows at the University’s Dairy Cattle Center. Krasno commented: “stop exploiting
animals. Get with the future and the future is consistent anti-oppression. Shut down the labs
and eat plants!” Within the same post, Krasno replied to another individual’s comment about
stopping animal experimentation and said: “I used to work in one of their labs. I made a bunch
of posts and tagged them in it and they hid all of them.” Krasno’s comment and reply were
hidden from view.9 However, the University did not hide a comment by user “rollietimothy”
who stated: “Hopefully I will be admitted next year, fingers crossed!” See dkt. 38-27.
On September 28, 2020, Krasno commented a second time on a University Instagram
post about the opening of a recreation center. She commented:
Thanks for continuing to delete my comments and untag yourself
from my photos. Definitely showcases fear. I will continue to

9

The parties do not dispute that at the time of the challenged conduct, there was no ability to
manually hide Instagram comments. Accordingly, I infer that Krasno’s September 2020 posts may have
been auto-moderated by the keyword filter, perhaps because she used the word “lab.”

12

share the truth about what it was like working in one of your
primate research labs and advocate for their closure. As I
mentioned before, today is a great day to shut down the primate
research labs!
At the time of her post, 179 other comments had been made to the same post. Krasno’s
comment was hidden from view.
On or shortly after that date, Moll imposed an account restriction on Krasno’s Instagram
account. Under the restriction, any comment posted from Krasno’s account on the University’s
Instagram page was automatically hidden from public view unless one of the University’s social
media managers manually “unhid” it. During the time the account was restricted, the University
still could see Krasno’s comments and Krasno still could see the University’s Instagram page,
posts, and comments.
By virtue of the account restriction, the following comments by Krasno on the
University’s Instagram account were hidden from public view:
•

On October 3, 2020, in response to a post by the
University about the birthday of its athletic team mascot
(Bucky Badger), Krasno commented: “close down the
primate labs! @uwmadison.”

•

On October 29, 2020, in response to a post by the
University about its efforts during the COVID-19
pandemic, Krasno commented: “@uwmadison stop testing
on monkeys!”

•

On November 17, 2020, in response to another post by the
University about COVID-19 measures, Krasno
commented: “[C]lose down the primate labs!
@uwmadison.”

•

On November 23, 2020, in response to a University post
about Thanksgiving travel, Krasno commented:
“@uwmadison, close down your primate research labs!”

13

•

On December 22, 2020, in response to a University post
about a dog being treated for cancer at the University’s
veterinary hospital, Krasno commented: “It is really quite
hypocritical the compassion shown to this dog while
thousands of animals languish in laboratories at
@uwmadison. I really wish you would acknowledge this
and do something about it.”

The University did not manually unhide any of these comments, even after the account-level
restriction was removed in January 2021.
On December 9, 2020, the University made a Facebook post that invited its Winter Class
of 2020 graduates to share their “#UWGrad pride” by “adding a frame” to their Facebook
profile photo. The post featured a digital “frame” depicting a photograph of a badger and the
words “PROUD#UWGRAD” that graduating students could superimpose on their own profile
picture. Dkt. 38-48. Krasno posted a comment that read: “University of Wisconsin-Madison,
are you really proud of all your graduates or just the ones who don’t object to your barbaric
treatment of monkeys in your research labs?” Dkt. 38-47. Moll saw this comment and manually
hid it from public view as being “off topic.” On the same comment thread, user Love Carolina
posted the word “Bravo,” and user Lynn Cooper posted “Emma Cooper.” Those posts were not
hidden or deleted.
On March 11, 2022, in response to a University Instagram post wishing its students a
safe and happy spring break, Krasno posted: “I bet the monkeys and other sentient beings in
your labs would like a break, or better yet, freedom.” Her post generated a number of replies,
including one from a user, seth_genteman, who stated in part: “Animals may end up saving your
life from the research conducted.” From there, Krasno and seth_genteman engaged in a series
of replies in the comment thread in which seth_genteman remarked, “[E]veryone would love an
14

easier way to obtain said results,” to which Krasno issued a final reply that referenced her time
in the University’s “primate lab.” Because Krasno’s final reply used the word “lab,” which was
on the keyword list, it was hidden.

V. Krasno’s Suit
Krasno sued the University in February 2021, alleging that by restricting her Instagram
account and deleting the single Facebook post she made in December 2020, the University was
engaging in unlawful censorship in violation of her First Amendment rights to free speech and
petition the government. By the time Krasno filed suit, the University had lifted its restriction
on Krasno’s Instagram account, but some of her comments were still being hidden by the
keyword filter.10 After Krasno discovered that the University’s keyword filter list was populated
by several words and phrases commonly used by her and other animal rights advocates, she
amended her complaint to allege that the keyword filters are viewpoint discriminatory and
burden her right to fully participate in the comment threads on the University’s social media
pages.
Krasno has at times been able to post her views in the comment threads by using creative
methods to avoid the keyword filter. For example, on May 9, 2021, Krasno wrote the following
comment on a University Instagram post discussing Mother’s Day: “What about all the mothers
you have in cages on campus? Celebrating them too after ripping their babies away from them?
A n l m a l t e s t l n g is cruel. [sic]” Similarly, Krasno commented on a University Facebook
post issued March 24, 2022, announcing a “Cool Science Image contest.” She commented:

10

It is unclear from the record whether defendants have manually hidden or deleted any of
Krasno’s comments to any post after Moll removed the account restriction in January 2021.

15

“Considering much of the ‘science’ done at the university is research without consent AKA t e
s t l n g o*n a n l m a l s, let’s see some footage the a l n m a l s in your l a b s. After all, y’all are
all about transparency, right? [all sic]” Both of these comments remain visible on the University’s
posts.
Krasno asserts official capacity claims for declaratory and injunctive relief against the
University’s chancellor and a vice chancellor, and official and individual capacity against Lucas,
Klein and Moll.11 In addition to damages, she seeks an injunction “requiring Defendants to
remove the restriction on Krasno’s ability to participate in comments on the University’s
Instagram and Facebook accounts.” Am. Complaint, dkt. 17, at 35.

OPINION
Summary judgment is appropriate when there are no genuine issues of material fact and
the moving party is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a). “The ordinary
standards for summary judgment remain unchanged on cross-motions for summary judgment:
we construe all facts and inferences arising from them in favor of the party against whom the
motion under consideration is made.” Blow v. Bijora, Inc., 855 F.3d 793, 797 (7th Cir. 2017).
If a party fails to make a showing sufficient to establish the existence of an element essential to
that party's case, and on which that party will bear the burden of proof at trial, then summary
judgment must be granted. Celotex Corp. v. Catrett, 477 U.S. 317, 322 (1986).

11

Krasno also sued the University’s Board of Regents, but she later dismissed that defendant
voluntarily. Dkt. 52, 57.

16

I. Free Speech Claims
The First Amendment provides in part that “Congress shall make no law . . . abridging
the freedom of speech . . .” U.S. Const. Amend. I. The First Amendment's freedom of speech
protection applies to states by virtue of the Fourteenth Amendment due process clause. Gitlow
v. New York, 268 U.S. 652 (1925); Edwards v. South Carolina, 372 U.S. 229, 235 (7th Cir. 1963).
Although social media platforms now are ubiquitous in America, courts only recently have
started grappling with when and how to apply developed First Amendment doctrine to these
relatively new and ever-evolving technologies. A number of courts have found that when the
government creates a social media page and invites users to leave comments or otherwise engage
with posted content by using the interactive features of the page, it creates a public forum in
which it may not discriminate against speakers based on viewpoint.

See, e.g., Garnier v.

O'Connor-Ratcliff, 41 F.4th 1158, 1185 (9th Cir. 2022) (school board trustees’ Facebook and
Twitter accounts were a public forum), pet. for cert. filed Oct. 6, 2022; Knight First Amendment Inst.
at Columbia Univ. v. Trump, 928 F.3d 226 (2d Cir. 2019) (finding President Donald Trump's
Twitter account to be a public forum, so that blocking users was unconstitutional viewpoint
discrimination); Davison v. Randall, 912 F.3d 666 (4th Cir. 2019) (public official violated First
Amendment by banning a critical constituent from a Facebook page); Robinson v. Hunt Cty.,
Texas, 921 F.3d 440 (5th Cir. 2019) (government official's act of blocking a constituent from
an official government social media page was unconstitutional viewpoint discrimination); Faison
v. Jones, 440 F. Supp. 3d 1123 (E.D. Cal. 2020) (granting plaintiffs’ motion for a preliminary
injunction and ordering defendant county sheriff to unblock plaintiffs on his official Facebook
page by finding the relevant page was a public forum); One Wisconsin Now v. Kremer, 354 F.

17

Supp. 3d 940, 949 (W.D. Wis. 2019) (state legislators’ Twitter accounts were designated public
fora ); Campbell v. Reisch, 367 F. Supp. 3d 987 (W.D. Mo. 2019) (denying motion to dismiss and
finding that defendant state legislator was acting under color of law when she blocked plaintiff
from her official Twitter account). Apart from this general rule, however, whether and to what
extent the government may establish and enforce content-based rules governing expression in
such online forums are questions that the courts are still sorting out.
In this case, the University concedes that the interactive portions of its social media pages
are public forums to which the First Amendment applies. It argues, however, that the interactive
comment sections, while open to the public, are limited for discussion of subjects related to the
underlying posts. Defendants assert that they hid Krasno’s posts because they were off-topic,
which they contend was a reasonable and viewpoint-neutral restriction in light of the purpose
of the forum. Krasno disagrees. As the parties urge, and as other courts have done, I look to the
Supreme Court’s forum jurisprudence to resolve this dispute.

A. Forum Analysis
The Court has described its forum jurisprudence as “a means of determining when the
Government's interest in limiting the use of its property to its intended purpose outweighs the
interest of those wishing to use the property for other purposes.” Cornelius v. NAACP Legal Def.
& Educ. Fund, Inc., 473 U.S. 788, 800 (1985). Under this approach, the court applies greater
or lesser scrutiny to the governmental restriction at issue depending on the nature of the forum.
John K. MacIver Inst. for Pub. Pol'y, Inc. v. Evers, 994 F.3d 602, 609 (7th Cir.), cert. denied, 142 S.
Ct. 711 (2021).

18

The Supreme Court has recognized three types of fora: (1) “traditional public fora,” such
as streets, sidewalks and parks; (2) “designated public fora,” which the government creates by
designating or opening a traditionally nonpublic forum for public discourse; and (3) “non-public
fora” which is a place the government has opened “only for specific purposes or subjects.”
Milestone v. City of Monroe, 665 F.3d 774, 783 n.3 (7th Cir. 2011).

In the first two

types,“speakers can be excluded . . . only when the exclusion is necessary to serve a compelling
state interest and the exclusion is narrowly drawn to achieve that interest.” Cornelius, 473 U.S.
at 800. “In a nonpublic forum, on the other hand—a space that ‘is not by tradition or
designation a forum for public communication’—the government has much more flexibility to
craft rules limiting speech.” Minnesota Voters All. v. Mansky, 138 S. Ct. 1876, 1885 (2018)
(quoting Perry Ed. Assn. v. Perry Local Educators' Assn., 460 U.S. 37, 46 (1983)); Cornelius, 473
U.S. at 799-800 (“Nothing in the Constitution requires the Government freely to grant access
to all who wish to exercise their right to free speech on every type of Government property
without regard to the nature of the property or to the disruption that might be caused by the
speaker's activities.”). The government may reserve such a forum “for its intended purposes,
communicative or otherwise, as long as the regulation on speech is reasonable and not an effort
to suppress expression merely because public officials oppose the speaker's view.” Mansky, 138
S. Ct. at 1885 (citing Perry, 460 U.S. at 46).
In other words, the term “nonpublic” does not mean a forum that it is closed to the public,
but rather one that is open to the public only for specified speakers or specified expressive
activities. Perry, 460 U.S. at 49 (“Implicit in the concept of the nonpublic forum is the right to
make distinctions in access on the basis of subject matter and speaker identity.”). Stated

19

another way, “a designated public forum is a nontraditional public space the Government has
opened to speech without restriction; a nonpublic forum is a nontraditional public space the
Government has opened to speech with restrictions.” Am. Freedom Def. Initiative v. Wash. Metro.
Area Transit Auth., 901 F.3d 356, 364 (D.C. Cir. 2018).
In recognition of this fact, courts often describe a forum opened by the government that
is limited to certain speakers or subjects as a “limited public forum.” Although most of these
cases use the term interchangeably with “nonpublic,” meaning that regulations in both are
subject to a lower level of scrutiny, at times the term “limited public forum” has been used to
describe a subcategory of “designated public” fora subject to the strict scrutiny test. See generally
Christian Legal Soc'y v. Walker, 453 F.3d 853, 866 n.2 (7th Cir. 2006) (describing confusion).
In the instant case, when the University argues that the comment threads to its social media
posts are “limited public” fora, I understand it to mean a forum governed by the reasonableness
and viewpoint neutrality requirements applied to “nonpublic” fora. To avoid confusion, I will
use the term “nonpublic” in this opinion to designate such a forum.

1. The Interactive Comment Threads are Nonpublic Fora
The parties agree that the relevant forum to be examined in this case is the interactive
comment threads hosted within the University’s posts on its @uwmadison Instagram account
and its @uwmadison Facebook page. They further agree that this forum is not a traditional
public forum. But they disagree whether the interactive comment section is a designated public
forum or a nonpublic forum. The University contends that the comment threads on its social
media pages are nonpublic forums because they are reserved for the discussion of specific topics,

20

namely, the topics of the underlying post the University makes. In other words, says the
University, it sets the agenda for discussion by posting about a specific topic; to the extent the
public is invited to participate, it is limited to discussing the topic of that particular post.
Krasno, on the other hand, contends that the comment threads are wide open for discussion of
anything related to the University, including its animal testing program. The resolution of this
dispute is potentially dispositive: the University does not contend that its moderation practices
would survive strict scrutiny if the comment threads are found to be designated public fora.
“The government does not create a public forum by inaction or by permitting limited
discourse, but only by intentionally opening a nontraditional forum for public discourse.”
Cornelius, 473 U.S. at 802. To determine that intent, courts must consider both explicit
expressions about intent and “the policy and practice of the government to ascertain whether
it intended to designate a place not traditionally open to assembly and debate as a public
forum.” Id. Courts also “examine[ ] the nature of the property and its compatibility with
expressive activity to discern the government's intent.”12 Id.
Beginning with the nature of the forum, there is no dispute that: the interactive
comment threads on the University’s Facebook and Instagram pages are expressly designed for
public commentary and debate; the University deliberately invites such commentary by leaving
the comment sections “open”; and, anyone with a Facebook or Instagram account has access to

12

The forum analysis is potentially complicated in this case because the University issued interim
guidance and modified its Social Media Statement after Krasno filed suit, which in theory may have
changed the boundaries of the forum. But neither party suggests that the court should sort the evidence
into “before” and “after” piles; both treat the forum designation as a singular question that includes
evidence post-dating the filing of this lawsuit. Although this could have complicated the analysis, it ends
up not mattering: as discussed infra, Krasno’s claims for relief based on the University’s pre-suit actions
are barred by state and qualified immunity no matter which forum category applies. Therefore, the court
need only conduct a forward-looking forum analysis.

21

the comment threads. However, the fact that the forum is generally open to all speakers and is
designed to facilitate expression doesn’t help answer the question whether particular expressive
activity – namely, off-topic comments – is inconsistent with the nature of the forum. Make The
Rd. by Walking, Inc. v. Turner, 378 F.3d 133, 144 (2d Cir. 2004) (“[T]he fact that ‘members of
the public are permitted freely to visit’ a forum . . . does not abrogate its nonpublic status if the
visitors are not permitted to express themselves freely in that forum.”)(citations and internal
quotations omitted). The Supreme Court has expressly recognized, for example, that a school
board may confine public discussion at its meetings “to specified subject matter[.]” City of
Madison, Joint Sch. Dist. No. 8 v. Wisconsin Emp. Rels. Comm'n, 429 U.S. 167, 176 (1976). The
question here is whether there is sufficient objective evidence to support the University’s
assertion that the public commentary it allows on its Instagram and Facebook pages has similar
subject matter limitations.
Thus, I turn to the inquiry whether the University has, “by policy or by practice,” limited
its forum to on-topic discussion, or whether it opened the comment threads “for indiscriminate
use by the general public.” Perry, 460 U.S. at 47. This inquiry “guards against the dangers of
post-hoc policy formulation or the discretionary enforcement of an effectively inoperative
policy.” Air Line Pilots Ass'n, Int'l v. Dep't of Aviation of City of Chicago, 45 F.3d 1144, 1153 (7th
Cir. 1995). In other words, the government’s stated policy during litigation is not dispositive.
In conducting this inquiry, a court may consider written policies, but it must pay more attention
to what the government actually does. Id. at 1154 (objective indicia of government intent more
telling than a “stated or paper policy”).

22

At the time this lawsuit was filed, the University’s Social Media Statement declared that
although the University did not regularly review content posted to its social media sites, it
reserved the right to remove content for a host of enumerated reasons, one of which was being
“off-topic.” The University subsequently removed the disclaimer, changed some of the language,
and added “not germane to the subject matter of the institutional post” as a reason that it may
remove content.
Other courts have found that government adoption and enforcement of rules of decorum
for public participation on its social media pages are inconsistent with an intent to create an
open forum for unlimited discussion. In Charudattan v. Darnell, No. 1:18CV109MW/GRJ, 2019
WL 12043587, at *6 (N.D. Fla. Feb. 7, 2019), for example, the court held that a Sheriff's
Office's Facebook page was a limited public forum because the “Sheriff's Office's content policy
limits the topics of discussion in response to the Sheriff's posts on that page[.]”. Similarly, in
Davison v. Plowman, 247 F. Supp. 3d 767, 776 (E.D. Va. 2017), aff'd, 715 F. App'x 298 (4th Cir.
2018), the court held that the defendant’s Social Media Comments Policy, which stated that
purpose of its Facebook page was “to present matters of public interest in Loudoun County” and
which reserved defendant’s right to remove comments deemed “clearly off topic,” established a
limited forum for purpose of discussion of the posted topics. The Ninth Circuit recently held
that the lack of any formal rules of decorum or etiquette was a factor supporting a finding that
the comment threads on school board trustees’ social media pages were designated public fora.
Garnier, 41 F.4th at 1185.
Here, although the University’s Social Media Statement does not specify that comments
must relate to the topics of its posts, its reservation of the right to remove content for

23

enumerated reasons signals that the University did not intend to open the interactive comment
threads for indiscriminate use. Rather, as in Charudattan and Plowman, the University’s Social
Media Statement indicates that the forum is open to public participation, subject to limitations.
The University’s retention of final authority over the kinds of speech activities it will allow is
inconsistent with an intent to create an unrestricted public forum. See Walker v. Tex. Div., Sons
of Confederate Veterans, Inc., 576 U.S. 200, 216 (2015) (exercise of final authority over speech
militates against a determination that the government intended to create a public forum).13
After Krasno filed this suit, the University issued Interim Guidance to its social media
moderators that purports to clarify how the moderators are to address off-topic content. As
Krasno points out, the guidance includes a number of confusing statements which suggest that
the University may consider a comment on topic if it relates to the particular social media page
on which the post is being made, rather than to the topic of the University’s post. While
admitting that its Interim Guidance is not a model of clarity, the University denies that it
interprets its off-topic rule this expansively and it maintains that the relevant comparison point
for application of the “off topic” prohibition is and always has been the content of the
University’s post. Moll 30(b)(6) Dep., dkt. 28, 70:13-19, 87:23-88:2; see also dkt. 54, ¶ 22. I
accept that contention. Cf. Mansky, 138 S. Ct. at 1889 (instructing that courts should consider
the government's “authoritative constructions” of their own law). Moreover, the University
appears to have taken Krasno’s critiques to heart when it amended its Social Media Statement

13

Krasno makes much of the fact that defendants testified that the University does not have a
Social Media “policy.” However, it is clear from their testimony that they were using the term “policy”
as a term of art within the University setting. All of the defendants identified the Social Media Statement
and the later-issued Interim Guidance as the relevant written guidance concerning content moderation.
Further, “[t]he absence of a formal policy does not prove the absence of a policy.” May v. EvansvilleVanderburgh Sch. Corp., 787 F.2d 1105, 1117 (7th Cir. 1986).

24

to reflect that it reserves the right to remove comments that are not germane to the subject
matter of the institutional post. Regardless, the salient point is that nothing in the Interim
Guidance suggests that the University intended to remove all restrictions on what users could
say on the comment threads and thereby open them up to unmoderated, indiscriminate use.
Krasno’s objections to the University’s policies are twofold. First, she says they do not
require its moderators to take action against off-topic comments. Second, contends Krasno, to
the extent that the University does moderate content, its efforts are so “hopelessly
underinclusive” and “slapdash” as to preclude a finding that the University’s social media pages
are subject to any meaningful limitations. In Krasno’s view, a wholly discretionary moderation
policy that fails to produce consistent results is, as a practical matter, no policy at all.
These are both valid points up to a point, but I do not find them persuasive. First,
“[s]electivity and discretion are some of the defining characteristics of the nonpublic forum.”
Griffin v. Sec'y of Veterans Affs., 288 F.3d 1309, 1323 (Fed. Cir. 2002). See also Cornelius, 473
U.S. at 804 (distinguishing nonpublic fora from those fora in which the decision to admit a
speaker is “merely ministerial”); Perry, 460 U.S. at 47 (that individual building principals had
to give permission for use of school mail systems tended to show that the mail system was a
nonpublic forum). As noted above, the University expressly invites the public to comment on
its posts within certain boundaries, and it reserves the right to remove content outside those
boundaries. This preservation of the right to decide for itself which speech to allow on its pages
is inconsistent with an intent to create an open forum for unfettered, unmoderated discussion.
Second, although inconsistent enforcement can support an inference that the government
intended to create a designated public forum, see, e.g., Garnier, 41 F. 4th at 1178; Air Line Pilots

25

Ass'n,, 45 F.3d at 1154, perfect enforcement is not required. New England Reg. Council of
Carpenters v. Kinton, 284 F.3d 9, 22 (1st Cir. 2002) (tolerance of some activities inconsistent
with the nonpublic nature of the forum not tantamount to affirmative act required to support
finding of designated public forum). In this case, I am not persuaded that the University’s
failure to moderate the comment threads with 100% consistency supports the inference that the
University intends to designate the comment threads open for indiscriminate expression. Put
the other way, the evidence establishes that University sufficiently attempts to moderate and
sufficiently does moderate the comment threads to qualify its Instagram and Facebook pages as
nonpublic fora.
First, there is no dispute that the University’s social media managers review thousands
of comments made in response to the University’s social media posts on its Instagram and
Facebook pages. In light of this, the examples submitted by each side fail to establish to any
degree of certainty what percentage of off-topic comments the University actually excludes.
What can be said with certainty is that the University applies its “off-topic” rule to exclude other
speech besides that pertaining to animal testing. This is plain from the keyword filter list as well
as the examples submitted by both defendants and Krasno.
Second, “First Amendment rights must always be applied ‘in light of the special
characteristics of the . . . environment’ in the particular case.” Healy v. James, 408 U.S. 169, 180
(1972) (quoting Tinker v. Des Moines Independent Community School District, 393 U.S. 503, 506
(1969)); United States v. Kokinda, 497 U.S. 720, 732 (1990) (plurality opinion) (“consideration
of a forum's special attributes is relevant to the constitutionality of a regulation since the
significance of the governmental interest must be assessed in light of the characteristic nature

26

and function of the particular forum involved.”) (quoting Heffron v. Int’l Soc. for Krisna
Consciousness, Inc., 452 U.S. 640, 650–651 (1981)). Precedent developed for the physical world
does not necessarily account for the nuances of how today’s online world operates. Packingham
v. North Carolina, 137 S. Ct. 1730, 1744 (2017) (Alito, J., concurring). The interactive comment
threads on social media pages are a unique type of fora, able to host a an almost unlimited
amount of expression by an unlimited number of unknown users. Depending on the topic, a
post by the University can generate thousands of comments, many arriving simultaneously or
in quick succession. Moreover, social media pages are “living documents,” meaning that after
the University generates a post, users may comment on that post (or remove their comments)
at any time, even months later, long after the topic’s social importance has faded and the
moderators no longer are focusing on that particular post. These factors, along with a policy of
lenity in borderline cases, unsurprisingly results in a number of arguably off-topic remarks
appearing in the comment threads. This result, however, does not convince me that the
University intended to allow off-topic comments as a matter of course.
Third, the Supreme Court has recognized that deference to government intent in
determining the nature of the forum may promote, rather than hinder, First Amendment
principles:
[W]e encourage the government to open its property to some
expressive activity in cases where, if faced with an all-or-nothing
choice, it might not open the property at all. That this distinction
turns on governmental intent does not render it unprotective of
speech. Rather, it reflects the reality that, with the exception of
traditional public fora, the government retains the choice of
whether to designate its property as a forum for specified classes of
speakers.
Arkansas Educ. Television Com’n v. Forbes, 523 U.S. 666, 680 (1998).
27

Here, the University has expressed an intent to open the comment threads to its social
media posts only to those speakers who have something to say that relates to the post, and it
claims to police the forum to the best of its ability. If the University were forced to choose
between (a) reviewing every single comment, with no margin for error, and (b) allowing anyone
to use the comment threads as a platform to speak about any subject at all, then it might well
choose not to open the comment threads at all, resulting in less speech, not more. I am not
persuaded that the Supreme Court’s forum precedents require the University to make this choice
between Scylla and Charybdis.
The cases cited by Krasno in which courts have found that the interactive portions on
social media sites were designated public forums are distinguishable. In Knight First Amendment
Inst. at Columbia Univ. v. Trump, 302 F. Supp. 3d 541, 575 (S.D.N.Y. 2018), the court held that
the interactive space of a tweet from the @realDonaldTrump Twitter account constituted a
designated public forum.14 Relying in part on the reasoning in Knight, this court reached the
same conclusion with respect to Twitter accounts created and maintained by state legislators,
finding the interactive spaces to be designated public fora that were subject to strict scrutiny.
One Wisconsin Now, 354 F. Supp. 3d at 949. In both of these cases, however, the courts were
considering the constitutionality of a ban that prevented a particular speaker from participating
in the forum at all. Whether the government actors could restrict the types of expression within
the forum was not at issue. Further, although several other courts have concluded that social
media pages (or their components) are “public” forums such that the First Amendment applied,

14

Although the Second Circuit later upheld the district court’s decision, that decision was vacated
as moot by the United States Supreme Court after President Biden was elected. See Knight First Amendment
Inst. at Columbia Univ. v. Trump, 928 F.3d 226 (2d Cir. 2019), cert. granted, judgment vacated sub nom.
Biden v. Knight First Amend. Inst. At Columbia Univ., 141 S. Ct. 1220 (2021).

28

most have done so without deciding what type of forum was created. See, e.g., Randall, 912 F.3d
at 687 (declining to decide whether defendant had created a public or non-public forum because
defendant’s ban of plaintiff amounted to viewpoint discrimination, which is “prohibited in all
forums.”) (citation omitted); Robinson, 921 F.3d at 447 (assuming for purposes of deciding
motion to dismiss that Sheriff’s Office page was a forum subject to First Amendment
protection); Czosnyka v. Gardiner, 2022 WL 407651, *2 (N.D. Ill. Feb. 10, 2022) (finding at
dismissal stage that interactive portions of government Facebook Page were a public forum);
Anderson v. Hansen, 519 F. Supp. 3d 457, 468 (E.D. Wis. 2021) (comment area of school
district’s Facebook page was public forum); Felts v. Reed, 504 F. Supp. 3d 978, 985 (E.D. Mo.
2020) (complaint fairly alleged that interactive component of public official’s Twitter account
was public forum); Faison, 440 F. Supp. 3d at 1135 (plaintiffs were likely to succeed in showing
that the interactive component of sheriff’s Facebook page, which defendant left open for public
discourse, was a public forum in which viewpoint discrimination was prohibited); Campbell, 367
F. Supp. 3d at 992 (public forum doctrine applies to state representative's Twitter Account, but
not classifying type of forum).
The Ninth Circuit in Garnier recently held that the social media accounts of two school
board trustees were designated public fora, but that case also is distinguishable. The school board
trustees in that case had “never adopted any formal rules of decorum or etiquette for their
pages,” and, until they began using extensive word filters that effectively prevented anyone from
commenting on their posts, their pages “were open and available to the public without any
restriction on the form or content of comments.” 41 F. 4th at 1178. Here, by contrast, the
University has a Social Media Statement that identifies certain types of unacceptable content,

29

including that deemed “off topic” and “not germane” to the institutional post, and it takes
measures to enforce the rules of the forum by manually and automatically hiding such content.
Apart from Charudattan and Plowman, cited above, the only other court to have
considered an “off-topic” rule in the context of the comment section of a social media page is
Kimsey v. City of Sammamish, 574 F. Supp. 3d 911 (W.D. Wash. 2021), a case cited by Krasno.
In that case, in considering the plaintiff’s motion for a preliminary injunction, the court found
that the comment section of a city’s Facebook posts was a designated public forum to which
strict scrutiny applied. Id. at 920. The court cited three reasons for its conclusion: (1) the city
had not limited access to its Facebook page but made it wide open to the public; (2) it enforced
its “off topic” rule inconsistently; and (3) the nature of Facebook itself and the enabled comment
field were designed and dedicated for expressive activities. Id. Respectfully, I do not find
Kimsey’s analysis persuasive. As noted above, simply because the government opens a forum
dedicated to expressive activity to any speaker does not prevent it from limiting the forum to
“specified subject matter.” Further, the Kimsey court’s emphasis on consistent enforcement does
not adequately account for the unique nature of the forum or the Supreme Court’s repeated
admonition that designating a public forum requires an affirmative act by the government, which
I have discussed above.
Planned Parenthood Ass'n/Chicago Area v. Chicago Transit Auth., 767 F.2d 1225 (7th Cir.
1985), another case cited by Krasno, is distinguishable. In that case, the court held that the
transit authority’s ad space was a designated forum where, apart from a “general contractual
directive . . . . to refuse vulgar, immoral, or disreputable advertising,” the authority maintained
no policy or system of control over ads it accepted, had no written guidelines, had accepted

30

controversial advertisements, and was found to have come up with such a policy solely to defend
its decision to reject plaintiff's advertising. Id. at 1232–33. The transit authority’s policing of
its ad space was so lax, found the court, that access was “virtually guaranteed to anyone willing
to pay the fee.” Id. at 1233. In the instant case, by contrast, access is not guaranteed:
comments containing certain words are hidden from the outset by the keyword filter, off-topic
comments that are not filtered stand a reasonable chance of being manually hidden or deleted,
and there is no evidence that the University came up with its “off topic” rule only to justify its
moderation decisions with respect to Krasno.
In sum, after examining the nature of the forum, the University’s written policies, and
its actual practices, I conclude that there is insufficient evidence to support Krasno’s contention
that the University routinely has allowed off-topic comments on its posts by everyone but
animal rights advocates like Krasno. To the contrary, the University has expressed an intent not
to open the comment threads on its social media posts to all posters for public dissemination of
their views on any topic whatsoever pertaining to the University. Regardless of the University’s
imperfect moderation efforts, it indisputably takes steps to hide or delete off-topic comments,
both those that relate to animal testing and those that do not. The fact that the University
moderates “loosely” and tolerates some arguably off-topic content – either by accident or by
design – may show that its restriction is unreasonable or viewpoint-based, but it does not show
that the University has invited the public to “use its facilities as a soapbox.” May, 787 F.2d at
1114. The comment threads are nonpublic fora.

31

2. The Off-Topic Rule is Reasonable and Viewpoint Neutral
Although there is no requirement of narrow tailoring in a nonpublic forum, the
government’s restrictions still must be viewpoint neutral and must be “reasonable in light of the
purpose served by the forum.” Mansky, 138 S. Ct. at 1886. In order to show that a speech
restriction is “reasonable,” the government must show that its restraint: (1) furthers a
“permissible objective;” and (2) contains “objective, workable standards” that are “capable of
reasoned application.” Mansky, 138 S. Ct. at 1886, 1891-92.
I conclude that the University has a legitimate, viewpoint-neutral interest in limiting the
comment threads to discussion of or reaction to the specific topic of the University’s post. The
University uses its Facebook page and Instagram account as channels to communicate official
University announcements, events and policies to the public, including its student body, and as
a means of promoting the UW-Madison “brand.” With respect to the interactive comment
threads, the University monitors what other social media users are saying in response to the
University’s posts, to see how its content is generally being received and to see the reactions its
posts are generating. The University also wants to see if anyone has questions, and it may
engage in its own speech in the comment threads to answer them. Allowing off-topic comments
to proliferate makes it more difficult for the University to engage with its followers and to see
comments to which it may wish to respond.
It also is legitimate for the University to consider the distraction that off-topic comments
may present to other users seeking to engage in and to discuss the topic of the University’s post.
It is reasonable for the University to conclude that these other users may be less inclined to leave
a comment, to ask a question, or to engage in on-topic discussion with other users if the

32

University’s pages are fraught with off-topic comments. As the Plowman court observed, “failure
to effectively moderate a public discussion may be as deleterious to dialogue in such a forum as
censorship.” 247 F. Supp. 3d at 778; see also Eichenlaub v. Twp. of Indiana, 385 F.3d 274, 281
(3d Cir. 2004) (noting that the First Amendment permits–and in some cases may require–a
government entity conducting a public meeting to stop a “speaker . . . try[ing] to hijack the
proceedings”).

There is nothing unreasonable about the University preferring that the

interactive comment threads have the look and feel of a brown bag lunch discussion rather than
its open-air Library Mall at the foot of State Street.15
Krasno argues that this court should find the University’s goal of preserving its comment
threads for on-topic discussion to be illegitimate because the University has not come forth with
evidence of a time when a large volume of off-topic comments that actually prevented it from
seeing a comment to which it would have responded, or with evidence that other users have
complained or stopped commenting because of a proliferation of such comments. Krasno further
points out that, unlike other public fora such as board meetings, where irrelevant commentary
can take up the board’s limited time for conducting business, Facebook and Instagram are

15

A cursory internet search reveals that among Big 10 universities, Northwestern, Penn State,
Minnesota, Rutgers, and Nebraska have rules against “off-topic” or “irrelevant” comments on their social
media pages. https://www.northwestern.edu/brand/applying-the-brand/social-media/social-media-guidelines/
https://hhd.psu.edu/faculty-staff/communications-and-marketing/social-media/community-guidelines-off
icial-penn-state , https://university-relations.umn.edu/resources/social-media-house-rules ,
https://communications.rutgers.edu/sites/default/files/0089_Social_Media_Guide_0628.pdf ,
https://nebraska.edu/offices-policies/external-relations/marketing-branding/social-media-community-gui
delines . Stanford, Princeton, and Berkeley have such rules as well.
https://ucomm.stanford.edu/policies/social-media-guidelines/, https://socialmedia.princeton.edu/guidelines#,
https://www.facebook.com/UCBerkeley/about_details. Although this is by no means a comprehensive or
even statistically significant survey, the University does not appear to be an outlier amongst its peers. Cf.
Mansky, 138 S. Ct. at 1888 (noting that “broadly shared judgment” that some kinds of campaign-related
clothing and accessories should stay outside polling place was factor showing legitimacy of Minnesota’s
objective).

33

designed to host dozens, if not hundreds of comments within a user’s posts. In light of this,
contends Krasno, off-topic comments are not inherently more disruptive of the purpose of the
forum than large amounts of on-topic comments, which the University indisputably tolerates.
I agree with Krasno that the University’s stated interests in limiting its fora to on-topic
speech are not so strong as to be unassailable, but in a nonpublic forum, they don’t have to be.
“In contrast to a public forum, a finding of strict incompatibility between the nature of the
speech or the identity of the speaker and the functioning of the nonpublic forum is not
mandated.” Cornelius, 473 U.S. at 808–09. Thus, whether large volumes of on-topic speech
may or may not be disruptive is not the question; the question is whether it is unreasonable for
the University to prohibit off-topic speech. Kokinda, 497 U.S. at 733 (“Whether or not the
[government] permits other forms of speech, which may or may not be disruptive, it is not
unreasonable to prohibit solicitation on the ground that it is unquestionably a particular form
of speech that is disruptive.”). Given that the University has a legitimate interest in hosting a
moderated forum for discussion of the subjects on which it posts, off-topic comments are, by
definition, more disruptive than on-topic comments.
It is true that in Garnier, the court found the trustees’ concerns about “visual clutter” and
distraction in their comment threads too weak to justify their complete blocking of the plaintiffs
from their Facebook and Twitter accounts absent proof of actual disruption. Id., 41 F. 4th at
1179-80. However, the court was applying the stricter scrutiny that applies to designated public
fora. Id., 41 F. 4th at 1179-80. In a nonpublic forum, proof of past disturbances is not required.
Perry, 460 U.S. at 52 n. 12; Cornelius, 473 U.S. at 810 (“the Government need not wait until
havoc is wreaked to restrict access to a nonpublic forum.”). What is more, the Garnier court

34

suggested that the trustees could address the annoyance caused by the plaintiffs’ repetitive
comments by either hiding or deleting the comments rather than blocking plaintiffs completely
or by “establish[ing] and enforc[ing] clear rules of etiquette for public comments on their pages,
including rules against lengthy, repetitive, or off-topic comments.” Id. at 1182 (emphasis
added). The University has done both here.
Having concluded that the University may hide or delete off-topic comments, the
remaining question to be answered is whether the University’s off-topic rule, which is
undoubtedly viewpoint neutral on its face, is “capable of reasoned application.” To meet this
test, “the State must be able to articulate some sensible basis for distinguishing what may come
in from what must stay out.” Mansky, 138 S. Ct. at 1888. This does not require eliminating all
discretion but merely that any discretion “must be guided by objective, workable standards.”
Id. at 1891. In Mansky, for example, the Court found that Minnesota’s law prohibiting anyone
from wearing a “political” badge, button or other political insignia at a polling place was too
vague to pass this test. Id. at 1888-89. Although the Court found that Minnesota had a
legitimate interest in maintaining a polling place free of partisan discord, Minnesota had failed
to offer any interpretations of the expansive term that were capable of reasoned application. For
example:
A shirt declaring “All Lives Matter,” we are told, could be
“perceived” as political. How about a shirt bearing the name of the
National Rifle Association? Definitely out. That said, a shirt
displaying a rainbow flag could be worn “unless there was an issue
on the ballot” that “related somehow . . . to gay rights.” A shirt
simply displaying the text of the Second Amendment? Prohibited.
But a shirt with the text of the First Amendment? “It would be
allowed.”
Id. at 1891 (emphasis in original).
35

Although the Court recognized that election judges screening individuals at the entrance
to the polls needed to have some degree of discretion and that “[p]erfect clarity and precise
guidance” were not required, the problems with Minnesota’s restriction went “beyond close calls
on borderline or fanciful cases” and was therefore unreasonable. Id.
Unlike a comment that is “political,” which can be evaluated on its own terms if the
operative definition is properly cabined, whether a statement is “off” or “on” topic is content
and context specific. To apply it, one needs an objectively sufficient understanding of the
substance and scope of the underlying topic. Even then, interpreting whether a comment is off
this topic necessarily will involve a fair amount of interpretive discretion, because “the point at
which speech becomes unduly repetitious or largely irrelevant is not mathematically
determinable.” White v. Norwalk, 900 F.3d 1421, 1426 (9th Cir. 1990).
Krasno contends that the subjectivity inherent in deciding whether something is off topic,
along with the undisputed evidence of inconsistent application, means that the University must
abandon the rule. See Mansky, 138 S. Ct. at 1891 (“It is ‘self-evident’ that an indeterminate
prohibition carries with it ‘[t]he opportunity for abuse, especially where [it] has received a
virtually open-ended interpretation.’”). However, Krasno hasn’t explained how the University
could preserve the forum for its intended use — discussion of the topics selected for posting by
the University – without vesting significant discretion in its moderators. Depending on the
nature of the forum, even a rule that “may defy objective description and may vary with
individual circumstances” is not necessarily unreasonable. Griffin, 288 F.3d at 1325 (rule vesting
discretion in VA administrators to ensure that cemeteries remain “sacred to the honor and

36

memory of those interred or memorialized there” was reasonable in light of characteristic nature
and function of national cemeteries).
Here, just like in any moderated discussion, a fair amount of judgment must be vested
in the moderator in order to ensure the forum serves its intended purpose. But that doesn’t
make the terms “not germane” or “off topic” wholly subject to the whims of the moderator. To
the contrary, although reasonable people may have different degrees of tolerance for when
something is “not germane” or “off topic,” the terms as commonly understood are sufficiently
objective to preclude wildly divergent applications, particularly now that the University has
made clear in its Social Media Statement that the comparison point for relevancy purposes is
the subject of the University’s post. Further, by prohibiting its moderators from engaging in
viewpoint discrimination, it has reduced the likelihood that the “off topic” rule will be used as
a cudgel to stifle speech with which the moderator disagrees. Cf. Southworth v. Bd. of Regents of
Univ. of Wisconsin Sys., 307 F.3d 566, 587-88 (7th Cir. 2002) (upholding regulations that
“express[ly] . . . prohibit[ ] viewpoint discrimination” and that require officials to “abide by the
principle of viewpoint neutrality”).
Finally, the existence of alternative channels of communication is a factor in the
reasonableness analysis. See Cornelius, 473 U.S. at 809; Perry, 460 U.S. at 53. Here, myriad
alternative means of communication exist by which Krasno, fellow animal rights advocates–and
everyone else in the world–may express their off-topic views about the University to the public.
To the extent the internet has become the “modern public square,” Krasno et omnes in mundo can
say whatever they wish about the University on their own media accounts, major, popular

37

platforms for which extend well beyond Facebook and Instagram.16 Given these alternatives and
the University’s professed intolerance of viewpoint discrimination, I am satisfied that the risk
that the University may sometimes hide arguably relevant comments does not outweigh its
interests in maintaining the comment threads for their intended purpose.
In sum, the University’s rule allowing for moderation of off-topic comments is a
reasonable and viewpoint neutral rule that furthers the University’s permissible interest in
preserving the interactive comment threads for discussion of the subjects posted by the
University. Krasno is free to post her views about testing on animals on her own pages or
anywhere else allowed on the internet. However, she has no First Amendment right to post
them on the University’s social media pages unless they are germane to the topic of the
University’s post.
It is against this backdrop that I now turn to the specific moderation decisions that
Krasno challenges in this case: (1) the restriction on her Instagram account from September
2020 to January 2021; (2) the manual deletion of a single Facebook comment she made in
December 2020; and (3) the University’s use of the keyword filter.

B.

Official Capacity Claims: Account Restriction and Deletion of Facebook Comment
The Eleventh Amendment bars suits by citizens against unconsenting states in federal

court. Seminole Tribe of Fla. v. Fla., 517 U.S. 44, 54 (1996). This amendment also protects arms
of the state, such as state agencies or state employees acting in their official capacity. Barnes v.

16

Although outmoded and of limited reach in the digital age, time-honored 20th Century messaging
techniques remain available: letters to the editors of print media, in-person rallies, printed flyers and
physical signs.

38

Bd. of Trustees of Univ. of Illinois, 946 F.3d 384, 391 (7th Cir. 2020); Indiana Prot. & Advocacy
Servs. v. Indiana Family & Soc. Servs. Admin., 603 F.3d 365, 370 (7th Cir. 2010).
Krasno argues that her official capacity claims fall within the exception carved out by Ex
parte Young, which holds that a plaintiff may proceed in federal court against a state official “for
the limited purpose of obtaining prospective relief against an ongoing violation of federal law.”
Driftless Area Land Conservancy v. Valcq, 16 F.4th 508, 518 (7th Cir. 2021), reh'g denied (Nov.
16, 2021). Where, however, a suit seeks to remedy only a past legal violation that has no
“ongoing” effects, it does not fall under the Ex parte Young exception to sovereign immunity. MCI
Telecommunications Corp. v. Illinois Bell Tel. Co., 222 F.3d 323, 345 (7th Cir. 2000) (“The plaintiff
must allege that the officers are acting in violation of federal law, and must seek prospective
relief to address an ongoing violation, not compensation or other retrospective relief for
violations past.”) (quoting Idaho v. Coeur d'Alene Tribe, 521 U.S. 261, 298-99 (1997) (Souter, J.,
dissenting)). See also Papasan v. Allain, 478 U.S. 265, 277–78 (1986) (The Ex parte Young
exception is “focused on cases in which a violation of federal law by a state official is ongoing as
opposed to cases in which federal law has been violated at one time or over a period of time in
the past.”).
Defendants argue that Krasno’s official capacity claim with respect to the Instagram
account restriction does not satisfy the Ex parte Young exception to state immunity because it was
lifted in January 2021. Krasno admits the restriction is no longer in effect, but she says the
violation is ongoing because the University has: (1) kept hidden the comments she made during
the four months it was in effect; and (2) remained silent about whether it might reimpose the
restriction in the future.

39

Neither of Krasno’s arguments is persuasive. First, although it may be technically true
that “social media never sleeps,” the fact remains that the online conversations in which Krasno
wanted to participate while her account was restricted are all but over. See Am. C.L. Union of
Illinois v. City of St. Charles, 794 F.2d 265, 274 (7th Cir. 1986) ([“S]peech delayed may be
speech destroyed; political speech . . . often is addressed to transitory issues, and becomes stale
when the issues pass away.”); Weinberger v. Romero-Barcelo, 456 U.S. 305, 311 (1982) (court
should not issue injunction “to restrain an act the injurious consequences of which are merely
trifling.”). Moreover, nothing prevents Krasno from making similar remarks again should they
be germane to a subject of a University post. She is not blocked or restricted from the forum.
Simply put, the violation with respect to the account restriction is not ongoing. That means
Young does not apply, and the Eleventh Amendment bars further proceedings on this claim in
federal court. As to Krasno’s second argument, the Eleventh Amendment bar applies, even
absent a finding of mootness. Watkins v. Blinzinger, 789 F.2d 474, 484 (7th Cir. 1986).
Accordingly, because the account restriction no longer is in effect and Krasno is not
suffering any ongoing harm from the University’s failure to “unhide” her old comments, her
official capacity claims for declaratory and injunctive relief based on the account-level restriction
is barred by the Eleventh Amendment and must be dismissed. (Although neither side addresses
it specifically, this holding applies to the deletion of Krasno’s single Facebook post in December
2020, as well.)

40

C. Official Capacity Claim: Use of Keyword Filter
This brings us to the University’s use of Meta’s keyword filter, which is the only ongoing
conduct for which Krasno could potentially obtain injunctive relief. Although Krasno has not
specified what kind of injunctive relief she is seeking, I presume she is seeking an order
prohibiting the University from using the keyword filter at all, or at least from populating its list
with words likely to be employed by users like her who oppose the use of animals for research.
Krasno argues that a mere glance at the list shows that the University is using the keyword filter
in a viewpoint-discriminatory way to target speech criticizing animal research. The University
denies this, claiming that the reason for the prevalence of anti-animal testing terms on the list
is because these terms appear repeatedly in off-topic spam campaigns, which happen to be
carried out most often by groups who share Krasno’s views about the University’s animal
research programs. The University would like to continue to use the filter on an as-needed basis
to defend against these (and other) spam-type campaigns that inundate the comment threads
with off-topic comments and make manual moderation difficult.
Although Krasno insists she has evidence to show that the University’s purported neutral
explanation for the keywords is a façade for viewpoint discrimination, I need not grapple with
this dispute because I find that Krasno lacks standing to seek an injunction against the
University’s use of the keyword filter. Schirmer v. Nagode, 621 F.3d 581, 584 (7th Cir. 2010)
(because standing is “an essential and unchanging part of the case-or-controversy requirement
of Article III” courts must consider it even if parties have not raised it).
The record shows only one instance in which Krasno made a comment that appears to
have been hidden by the keyword filter and that was arguably on topic to the subject of the

41

University’s post: her “stop exploiting animals” comment in response to the University’s
September 2020 post about cows returning to the University’s Dairy Cattle Center. But the
mere fact that Krasno may have suffered past harm for which she has standing to seek damages
“does not necessarily carry over” to her facial challenge requesting an injunction against any use
of the keyword filter by the University: “A plaintiff ‘must demonstrate standing separately for
each form of relief sought.’” Schirmer, 621 F.3d at 585 (citations omitted). See also City of Los
Angeles v. Lyons, 461 U.S. 95, 109 (1983) (plaintiff had standing to seek damages but not
injunctive relief against abusive police practices).
To establish standing to seek prospective relief against the University’s use of the
keyword filter, Krasno must show that: (1) she is under threat of an actual and imminent injury
in fact; (2) there is a causal relation between that injury and the conduct to be enjoined; and (3)
it is likely, rather than speculative or hypothetical, that a favorable judicial decision will prevent
or redress that injury. Id. (citations omitted).
Krasno cannot satisfy the first or third requirements. She asserts the keyword filters
impose “ongoing injury” because they cause her to “alter her messaging” against animal research
to avoid further censorship. Br. in Supp., dkt. 35, at 58. As evidence, she cites to her May 9,
2021, comment in response to the University’s post about Mother’s Day and to her March 24,
2022, comment in response to the University’s announcement of its “Cool Science Image
contest,” in which she modified the spelling and spacing of certain words to avoid the keyword
filter.

However, implicit in Krasno’s “harm” argument is her belief that she has a

constitutionally-guaranteed right to comment about animal testing regardless of the topic of the
University’s posts. I already have found that she doesn’t. Neither of these comments was

42

germane to the University’s posts, regardless whether the University opted not to manually
delete them.17 Accordingly, she was not harmed by the keyword filter in these instances.
Nor has Krasno shown that she is likely to be injured by the keyword filter in the future.
The keywords will have a viewpoint-discriminatory effect on Krasno’s ability to speak out against
the research the University conducts on animals only if the University makes a post on some
topic to which Krasno’s views about animal testing are germane. Even then, whether the
keyword list will result in her comments being hidden is speculative, because the University
changes the words on an as-needed basis, and none of us knows how Krasno will phrase her
input. Finally, future harm is reduced nearly to the vanishing point by the University’s assertion
that, if it posts about animal research, then it will remove pertinent key words from the list so
that on-point posts will not be blocked.
Krasno argues that the University should not be taken at its word because it has posted
about animal research in the past yet left its keywords in place. This is a valid criticism. As
Krasno points out, the first post she commented on, in September 2020, was about cows
returning to the University’s Dairy Cattle Center, yet her “stop exploiting animals” comment
was hidden because she used the word “labs.” She also notes that the University posted in
December 2020 about veterinary care provided to a dog, but it hid her comment that it was
“really quite hypocritical the compassion shown to this dog while thousands of animals languish
in laboratories at @uwmadison.” Although this second comment was hidden by virtue of the
account restriction and not the keyword filter, the salient point is that the University has in the

17

The only other time the keyword filter allegedly interfered with Krasnow’s speech was when her
final response to user seth_genteman was hidden, but there, too, Krasno was speaking off-topic:
seth_genteman had posted a response to Krasno’s initial off-topic comments about the University’s animal
research program in response to the University’s March 11, 2022, post wishing its students a safe and
happy spring break.

43

past featured animals in its posts. The University denies in its brief that these posts concerned
“animal testing in the first instance,” dkt. 33, at 32, but this tight concept of relevancy seems
at odds with the actual approach taken by its moderators: Moll has since conceded that
Krasno’s December 2020 comment was on topic.
Nevertheless, to find that a plaintiff has standing to seek injunctive relief, the court must
be satisfied that the plaintiff faces a “realistic threat” that the past harmful events will be
repeated. Lyons, 461 U.S. at 102 (citing Rizzo v. Goode, 423 U.S. 362 (1976)); see also Bryant v.
Cheney, 924 F.2d 525, 529 (4th Cir. 1991) (injunctive powers of the federal courts are broad,
but “Article III simply precludes their empty use to enjoin the conjectural or declare the fully
repaired broken.”).

“Article III of the Constitution bars a federal court from enjoining

threatened action that the plaintiff has no reason to suppose even remotely likely ever to
materialize; there must be a real dispute in the sense that its resolution is likely to have tangible
consequences for the plaintiff.” Lawson v. Hill, 368 F.3d 955, 957 (7th Cir. 2004).
That threshold has not been met in this case. Krasno has shown only a handful of
isolated instances in which she or others made an arguably relevant comment that was, or
possibly would have been, hidden by the keyword filter, and these instances appear to have
occurred before Krasno filed this lawsuit. Since then, the University has taken steps to clarify
its moderation practices and it now directs its moderators not to remove comments merely
because they are disagreeable. The University also has changed the words on the keyword filter,
and it may have changed them again after the parties filed their summary judgment submissions.
As Krasno concedes, the process of adding or removing words from the keyword filter is

44

“situation dependent.” Finally, the University has the ability to manually “unhide” comments
that it would otherwise deem on-topic but for the keyword filter.
The upshot is that the likelihood that Krasno will make on-topic comments that are
hidden by the keyword filter depends on variables that are entirely conjectural at this point,
namely, the content of the University’s post, the content of the keyword list, and the content
of Krasno’s comment. The risk of harm to Krasno is simply too speculative and remote to confer
upon her Article III standing to enjoin the University from using the keyword filter to defend
itself against off-topic spam attacks. Finally, even if standing were present, I nevertheless would
decline to enter an injunction based merely on the ability to hypothesize scenarios in which the
filters might have a viewpoint discriminatory effect on Krasno or others not party to this lawsuit.
Lawson, 368 F.3d at 960 (“courts do retain the authority to deny an injunction even if a wrong
is proved and a threatened harm shown”). Accordingly, Krasno’s official capacity claim with
respect to the keyword filter will be dismissed.

D. Individual Capacity Claims
Krasno asserts that there is sufficient evidence in the record to show that the individual
defendants, Lucas, Klein, and Moll, restricted her account and hid some of her comments not
because her comments were off-topic, but because they did not like her views against animal
testing. Defendants deny this, but they also have asserted a qualified immunity defense, which
protects government officials from liability if “their conduct does not violate clearly established

45

statutory or constitutional rights of which a reasonable person would have known.”18 Pearson
v. Callahan, 555 U.S. 223, 231 (2009) (quoting Harlow v. Fitzgerald, 457 U.S. 800, 818 (1982)).
Defendants’ assertion of qualified immunity means that, in addition to showing a
violation of a constitutional right, Krasno must show that the constitutional right was clearly
established at the time of the alleged violations. Leiser v. Kloth, 933 F.3d 696, 701 (7th Cir.
2019). The court may address these prongs in any order. Tolan v. Cotton, 572 U.S. 650, 656
(2014). Because I conclude that Krasno has not met her burden on the second prong, I need
not decide whether viewpoint discrimination occurred in the first instance.
In conducting a qualified immunity analysis, the inquiry “must be undertaken in light
of the specific context of the case, not as a broad general proposition.” Brosseau v. Haugen, 543
U.S. 194, 198, 125 (2004) (per curiam) (internal quotation marks omitted). Thus, it is not
enough that a rule is suggested by existing precedent; the “rule’s contours must be so well
defined that it is ‘clear to a reasonable officer that his conduct was unlawful in the situation he
confronted.’” District of Columbia v. Wesby, 138 S. Ct. 577, 590 (2018). In other words, to
claim–as Krasno essentially does–that “the law prohibits government officials from engaging in
viewpoint discrimination in a public forum” is not specific enough.
Instead, to defeat the individual defendants’ claim of qualified immunity, Krasno must
come forth with precedent clearly establishing that the interactive comment section on a
government-controlled social media site is a public forum subject to the First Amendment. See
Garnier, 41 F. 4th at 1183 (defendants were entitled to qualified immunity because it was not

18

Klein and Lucas have also denied being personally responsible for any of the challenged
moderation decisions. I need not decide this issue in light of my finding on qualified immunity.

46

clearly established in 2017 that the Garniers had a “First Amendment right to post comments
on a public official's Facebook or Twitter page.”); Siddique v. Laliberte, 972 F.3d 898, 904–05
(7th Cir. 2020) (granting qualified immunity where plaintiff did not identify “any analogous
comparators applying First Amendment retaliation principles to student-government
operations”); Surita v. Hyde, 665 F.3d 860, 872 (7th Cir. 2011) (rejecting qualified immunity
defense where “[b]efore 2004 several cases applied [forum analysis] to settings of public hearings
and meetings of government bodies.”). The court looks first to controlling precedent on the
issue from the Supreme Court and then to precedent from this circuit. Estate of Escobedo v.
Bender, 600 F.3d 770, 781 (7th Cir. 2010).
Krasno begins by citing Packingham v. North Carolina, 137 S. Ct. 1730 (2017), which
declared social media to be one of “the most important places (in a spatial sense) for the
exchange of views.” Id. at 1735. However, the Court in that case was considering only whether
North Carolina could make it a felony for a registered sex offender to gain access to social media
sites, not whether the comment threads of a government-controlled social media site constitute
public fora. Thus, regardless of the Court’s sweeping observations about the importance of social
media to public discourse, Packingham did not resolve any questions pertaining to government
moderation of its own social media sites. Likewise, the Seventh Circuit has not addressed this
question.
In the absence of any analogous Supreme Court or Seventh Circuit cases, the analysis
shifts to whether the clearly established weight of authority from other courts would have put
every reasonable official on notice beginning in September 2020 that it is unlawful to engage in
viewpoint discrimination with respect to comments left by users on the comment threads of a

47

government-controlled social media site. Krasno cites the Fourth Circuit’s decision in Davison
v. Randall, 912 F.3d 666 (4th Cir. 2019), which was the first circuit court of appeals to find that
the interactive spaces of a social media page was a public forum in which the First Amendment
applied. She also cites the Second Circuit’s decision in Knight First Amendment Institute at
Columbia University v. Trump, 928 F.3d 226, 237 (2d Cir. 2019), in which the court held that
President Trump created a “public forum” when he “repeatedly used the [presidential Twitter]
Account as an official vehicle for governance and made its interactive features accessible to the
public without limitation.” However, as noted above, after President Biden was elected, the
Supreme Court granted certiorari, vacated the Second Circuit’s judgment, and sent the case back
with instructions to dismiss it as moot. Biden v. Knight First Amend. Inst. at Columbia Univ., 141
S. Ct. 1220 (2021). Accordingly, Knight offers at most “persuasive authority,” see Camreta v.
Greene, 563 U.S. 692, 713 (2011) (vacatur “strips the decision below of its binding effect”)
(citations omitted).19
In any event, neither Davison nor Knight is a Seventh Circuit decision. As noted above,
although by now it seems reasonably clear that the First Amendment’s requirement of viewpoint
neutrality applies to the interactive comment threads of a government-controlled social media
site, Krasno cites only three federal cases pre-dating the University’s conduct that reached that
conclusion: One Wisconsin Now, 354 F. Supp. 3d 940, 953–55 (W.D. Wis. 2019), Faison v. Jones,
440 F. Supp. 3d 1123, 1135 (E.D. Cal. 2020), and the district court’s decision in Garnier, 2019

19

Krasno points out that Moll testified that he was aware of the Knight case at least as early as
April 2020 and he understood it as broadly applying to the University’s action of “blocking users” on
“social media in general.” However, Krasno does not cite and I am not aware of any authority for her
suggestion that a defendant’s own interpretation of a non-binding court decision defeats his subsequent
claim of qualified immunity made by his attorneys.

48

WL 4736208, at *10-11 (S.D. Cal. Sept. 26, 2019). But district court decisions – even from
this court – cannot create clearly established law. Camreta, 563 U.S. at 709 n.7. All told, the
handful of cases cited by Krasno did not amount to a “clear trend” prohibiting viewpoint
discrimination on government-controlled social media sites in late 2020-early 2021.
In sum, even assuming, arguendo, that Krasno could establish that defendants
discriminated against her viewpoint when they restricted her Instagram account or otherwise hid
or deleted her comments on the University’s social media pages, there was no clearly established
precedent placing the unlawfulness of that conduct beyond debate at the time those decisions
were made. Accordingly, the individual defendants are entitled to qualified immunity.

II. Right to Petition for Redress of Grievances
Finally, defendants argue that Krasno lacks any viable claim that defendants violated her
right to petition the government for redress of grievances.

I agree.

Although the First

Amendment undoubtedly guarantees citizens the right to “express their ideas, hopes, and
concerns to their government,” it does not require the government “to listen or respond to
individuals’ communications on public issues.” Shipley v. Chicago Bd. of Election Comm., 947 F.3d
1056, 1063 (7th Cir. 2010) (citations omitted). In Shipley, the plaintiffs alleged that the election
board had violated their right to petition their government by preventing them from publicly
commenting at a meeting before the Board certified precinct election returns. Upholding the
district court’s dismissal of the claim, the Seventh Circuit noted that plaintiffs’ own allegations
showed that they were able to voice their objections to the Board on other occasions, including
in writing in advance of the meeting. Id. Because plaintiffs’ complaint was only that “they were

49

not able to petition the Board at their desired time and place, not that they were prohibited from
petitioning the government,” they failed to state a claim. Id. (emphasis added).
The same goes for Krasno. As she concedes, she has multiple other ways besides posting
on the comment threads of the University’s social media accounts to express to the University
her concerns about animal testing; indeed she has employed some of them. And Krasno fails to
cite any persuasive, much less controlling, authority for her position that the existence of these
alternative means does not doom her claim. Accordingly, defendants are entitled to summary
judgment on this claim.

ORDER
IT IS ORDERED that:
1. Plaintiff Madeline Krasno’s motion for summary judgment, dkt. 34, is DENIED;
2. Defendants’ motion for summary judgment, dkt. 31, is GRANTED.
3. The clerk of court is directed to enter judgment for defendants and close this case.

Entered this 2nd day of November, 2022.
BY THE COURT:
/s/
_______________________
STEPHEN L. CROCKER
Magistrate Judge

50