Introduction

Nick Emmett was a high school senior in 2000.¹Emmett v. Kent Sch. Dist. No. 415, 92 F. Supp. 2d 1088, 1089 (W.D. Wash. 2000). He was a good student with a 3.95 GPA; he was co-captain of his basketball team; and he had no disciplinary history with his school.²Id. On February 13, 2000, Nick decided to create a web page titled the “Unofficial Kentlake High Home Page.”³Id. He was at home, using his own free time.⁴Id. The site included disclaimers warning visitors that his page was neither sponsored nor approved by Kentlake High School and that the site was purely for entertainment.⁵Id. The site had a section with mock “obituaries” for students at the high school.⁶Id. Visitors to the site could vote who would “die” next—meaning, who would get the next obituary written for them.⁷Emmett, 92 F. Supp. at 1089. These “obituaries” were “written tongue-in-cheek,” and they became a joke that students sometimes discussed at school.⁸Id.

Unfortunately for Nick, an evening news story characterized his site as a “hit list.”⁹Id. Based on the news story, the principal summoned Nick to his office and informed Nick that he would be expelled for “intimidation, harassment, disruption to the educational process, and violation of Kent School District copyright.”¹⁰Id. The expulsion was later changed to a five-day out-of-school suspension, but the suspension included a prohibition on participating in school sports.¹¹Id. Nick would not be allowed to play in his basketball team’s playoff game even though he was co-captain.¹²Id. The district presented neither evidence that students felt threatened or intimidated by Nick’s words, nor evidence that Nick had intended the “obituaries” as a threat.¹³Emmett, 92 F. Supp. at 1089.

Aaron Wisniewski’s online joke was much darker than Nick’s. Like Nick, Aaron used a home computer to create and disseminate his internet speech.¹⁴Wisniewski ex rel. Wisniewski v. Bd. of Educ., 494 F.3d 34, 36 (2d Cir. 2007). Near the end of eighth grade, Aaron instant messaged fifteen of his friends, some of whom were schoolmates, with a drawing of “a pistol firing a bullet at a person’s head, above which were dots representing splattered blood,” beneath which were the words “Kill Mr. VanderMolen.”¹⁵Id. Mr. VanderMolen was Aaron’s English teacher.¹⁶Id. One of Aaron’s schoolmates showed the drawing to Mr. VanderMolen, who forwarded the drawing to both the high school and middle school principals, who in turn brought the matter to the attention of the superintendent and the police.¹⁷Id. When he was questioned by the school authorities, Aaron admitted to creating the drawing and expressed regret.¹⁸Id. He was suspended for five days pending a superintendent’s hearing.¹⁹Id.

The police concluded that Aaron had meant the drawing as a joke, he now understood the severity of his actions, and he posed no threat to the school or to Mr. VanderMolen.²⁰Wisniewski, 494 F.3d at 36. A psychologist also evaluated Aaron and found that Aaron “had no violent intent, posed no actual threat, and made the icon as a joke.”²¹Id. But at the superintendent’s hearing, the designated hearing officer decided that the drawing was “threatening and should not have been understood as a joke.”²²Id. She acknowledged the opinions of the police and the psychologist but decided that Aaron’s “intent [was] irrelevant.”²³Id. Aaron was then suspended for a full semester, and due to “school and community hostility,” the family moved cities.²⁴Id. at 37.

Fortunately, the court granted Nick Emmett relief, enjoining the school from enforcing the suspension.²⁵Emmett v. Kent Sch. Dist. No. 415, 92 F. Supp. 1088, 1090 (W.D. Wash. 2000). But the courts upheld the school suspension against Aaron Wisniewski.²⁶Wisniewski, 494 F.3d at 37–38. However distasteful the jokes, it still seems wrong that students like Nick and Aaron should have to appeal to a court to prevent or be compensated for school punishment aimed at their internet speech.

One might wonder why the schools could punish Nick and Aaron in the first place, especially given the finding in both cases that the students’ messages were jokes and not real threats. Courts have given schools little guidance and broad discretion when dealing with student internet speech. The Supreme Court articulated a rule for school regulation of on-campus speech in a line of cases beginning with Tinker v. Des Moines Independent Community School District²⁷393 U.S. 503 (1969). and ending with Morse v. Frederick.²⁸551 U.S. 393 (2007). However, none of those cases dealt with internet speech, and none of them dealt with off-campus speech.²⁹The Court ruled that the speech in Morse fit within the on-campus category because it occurred at a student-sponsored event even though not properly on school grounds. 551 U.S. at 408–10. The circuit courts of appeals have been left to decide how the Court’s precedents control school authority over off-campus internet speech, and the result has been confusion.³⁰See discussion infra Parts II.A–C. There are nearly as many tests as there are circuits, and some courts, like the Third Circuit, have refused to decide the question definitively.³¹See discussion infra Part II.C.

The main commonality of the tests articulated by the circuit courts is the heavy reliance on the speech’s effects at school.³²See infra notes 153–57 and accompanying text. Almost uniformly, the courts have dismissed the on-campus versus off-campus distinction as having a territorial meaning in relation to internet speech.³³See infra notes 153–57 and accompanying text. However, Tinker strongly suggests that the territorial part of the on-campus versus off-campus distinction is an important check on the scope of school authority over student speech.³⁴See Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 506, 509–14 (1969).

One reason for the confusion among the courts is that their tests ask how much internet speech must affect the school and its administration to allow the school to regulate it, a metric that is imprecise and impossible to quantify. This Comment proposes turning from the futile attempts to draw a more precise effects test to resurrecting the territorial nexus check on school authority first established by Tinker. Schools should be prohibited from punishing students for their speech on the internet entirely, unless that speech was on a school-owned website. Schools can bar access to certain sites from school-provided machines or internet connections to maintain order while students are at school,³⁵See infra notes 177–81 and accompanying text. but they should not be able to punish students for speech posted on internet sites owned publicly or by other private entities. This solution would provide a clear rule for schools to follow and constrain their authority within the limits set by the Supreme Court in Tinker and its progeny.

Part I of this Comment discusses the Supreme Court precedents that form the framework for school authority to regulate student speech. Part II explores the circuit split over the issue of school regulation of student internet speech. To limit its scope, this Comment will primarily consider cases that deal with internet speech, though the courts of appeals’ confusion affects other forms of off-campus speech as well. Part II also discusses the flaws in the courts’ reasoning that raise concerns for protecting students’ rights. Finally, Part III defends the two-part solution of resurrecting the territorial nexus for the scope of the school’s power and prohibiting schools from punishing internet speech itself, limiting schools’ disciplinary power to only expressive conduct that occurs on campus and causes substantial disruption.

I.     Background—Supreme Court Precedent

Students in primary and secondary schools have First Amendment rights, and Tinker is a bold champion of these rights.³⁶See Tinker, 393 U.S. at 506. In Tinker, the Court upheld the rights of students to express their political views, even while at school.³⁷The Court ruled that the school had specifically targeted the black armbands (which expressed disapproval for the Vietnam War) based on viewpoint, and that because the students were not violating a dress code or causing disturbances, they should not have been punished for wearing the armbands. Id. at 507–14. However, Tinker is best known for articulating a standard for restricting student speech.³⁸Id. at 513. The Court wrote:

When [the student] is in the cafeteria, or on the playing field, or on the campus during the authorized hours, he may express his opinions, even on controversial subjects . . . if he does so without “materially and substantially interfer[ing] with the requirements of appropriate discipline in the operation of the school” and without colliding with the rights of others.³⁹Id. at 512–13 (quoting from Burnside v. Byars, 363 F.2d 744, 749 (5th Cir. 1966)).

This language reads as a writ of liberty to the student—that he has freedom to express his ideas at school, even if not within the strict confines of a classroom—and the Supreme Court has often interpreted it this way.⁴⁰See, e.g., Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 266 (1988). But most, including the circuit courts of appeals,⁴¹See, e.g., Bell v. Itawamba Cty. Sch. Bd., 799 F.3d 379, 390 (5th Cir. 2015) (en banc); Wynar v. Douglas Cty. Sch. Dist., 728 F.3d 1062, 1067–69 (9th Cir. 2013); Kowalski v. Berkeley Cty. Sch., 652 F.3d 565, 571–72 (4th Cir. 2011); D.J.M. v. Hannibal Pub. Sch. Dist., 647 F.3d 754, 765 (8th Cir. 2011); Doninger v. Niehoff, 527 F.3d 41, 50 (2d Cir. 2008). have focused on the “materially and substantially interfering” standard as a limitation on this freedom, alongside the succeeding sentence of the opinion:

But conduct by the student, in class or out of it, which for any reason—whether it stems from time, place, or type of behavior—materially disrupts classwork or involves substantial disorder or invasion of the rights of others is, of course, not immunized by the constitutional guarantee of freedom of speech.⁴²Tinker, 393 U.S. at 513.

In context, the Tinker limitations of “materially and substantially interfering,” “materially disrupts,” and “involves substantial disorder” all refer to activity on the school campus, though not always within the classroom proper.⁴³Id. at 509–14. The Court has upheld this territorial distinction in subsequent cases. In Hazelwood School District v. Kuhlmeier,⁴⁴484 U.S. 260 (1988). the Court stated, “A school need not tolerate student speech that is inconsistent with its ‘basic educational mission,’ even though the government could not censor similar speech outside the school.”⁴⁵Id. at 266 (quoting Bethel Sch. Dist. No. 403 v. Fraser, 478 U.S. 675, 685 (1986)). The rights of students while at school are “applied in light of the special characteristics of the school environment.”⁴⁶Id. (quoting Tinker, 393 U.S. at 506). These statements, along with the context of the case,⁴⁷The issue in Hazelwood was whether a high school could refuse to print certain articles in the school paper because of various academic and journalistic concerns, such as protecting the anonymity of the student subjects of the article (the refusal was not solely based on content). The Court ruled that because the paper was almost entirely controlled by the teacher and the principal, it was a school-sponsored publication, not a public forum, and the school could exercise greater control over it than might be exercised over actual newspapers. See id. at 270–72. imply that schools can regulate speech that is inside the school, but that power does not extend outside of the school.⁴⁸See id. at 266.

Similarly, the Court in Bethel School District No. 403 v. Fraser⁴⁹478 U.S. 675 (1986). distinguished between restrictions a student might face in the greater society and the control appropriately exercised over him by a school.⁵⁰See id. at 680, 682. The Court reaffirmed the statement in Tinker that students still retain their First Amendment rights and “do not shed [them] at the schoolhouse gate.”⁵¹Tinker, 393 U.S. at 506. The Court then stated that a school could tell a student not to use lewd, graphically sexual speech during a class officer campaign and could punish him if he used such language despite the warning.⁵²Fraser, 478 U.S. at 680–82. However, nothing in Fraser could be construed as saying that the school could punish the student for saying the same things if, for instance, he chose to say them to his associates in a public park. Rather, the holding in Fraser rests solely on the speech’s location at the school and occurrence during a school event.⁵³See id. at 680–81.

A more recent Supreme Court decision, Morse v. Frederick,⁵⁴551 U.S. 393 (2007). slightly widened the scope of “on-campus” or “at school.” The Court in Morse ruled that events that are sanctioned and supervised by the school count as being “at school,” even if those events are not actually on school property.⁵⁵Id. at 401. The student in this case had displayed a banner with the text “Bong Hits 4 Jesus” while observing the 2004 Olympic Torch Relay pass through his hometown; this participation was permitted as an “approved social event or class trip” since the relay passed during school hours. Id. at 397. Thus, the Court found that the event was “school-sanctioned” or a “school event.” See id. at 400–01, 403. However, the Court remained clear that being “at school” is the necessary link to trigger any school regulation of student speech.⁵⁶See id. at 396–97, 400–01. Moreover, the speech itself must be made “at school” to be censored or restrained by the school; students’ rights outside of school remain unaffected by Tinker principles.⁵⁷See id. at 396–97, 401; see also Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 271 (1988); Porter v. Ascension Par. Sch. Bd., 393 F.3d 608, 615 (5th Cir. 2004).

II.     Problems with Regulating Internet Speech

The circuit courts of appeals’ decisions fall into three major groups.⁵⁸See discussion infra Parts II.A., II.B., and II.C., respectively. Note that the First, Sixth, Seventh, Tenth, and D.C. Circuits have not yet considered Tinker’s application to off-campus internet speech. For this reason, this Comment does not discuss cases from these circuits. The first group (the Second, Fourth, Eighth, Ninth, and Eleventh Circuits) has abandoned the territorial nexus component of Tinker as impossible, often without much more than a nod to that component’s existence. Instead, this group has given the schools broad authority and discretion to discipline off-campus speech, relying solely on the substantial disruption test to determine whether a student’s rights are violated. The second group (the Fifth Circuit) has also abandoned the territorial nexus requirement and substituted an intent to reach the school community for the traditional physical location test. The intent requirement is supposed to work alongside the substantial disruption test as a check on school authority. The third group (the Third Circuit) is still unsure of the proper relationship between school authority and off-campus speech but recognizes that the territorial nexus requirement is important to check the school’s power. No test has been formulated to incorporate this recognition in the third group.

A.     Group 1: Territorial Nexus Does Not Matter; The Schools May Discipline Off-Campus Speech at Their Discretion if a Substantial Disruption Is Reasonably Foreseeable

The Second,⁵⁹See Doninger v. Niehoff, 527 F.3d 41, 48 (2d Cir. 2008). Fourth,⁶⁰See Kowalski v. Berkeley Cty. Sch., 652 F.3d 565, 567 (4th Cir. 2011). Eighth,⁶¹See D.J.M. v. Hannibal Pub. Sch. Dist., 647 F.3d 754, 764–65 (8th Cir. 2011). Ninth,⁶²See Wynar v. Douglas Cty. Sch. Dist., 728 F.3d 1062, 1069 (9th Cir. 2013). and Eleventh⁶³See Doe v. Valencia Coll., 903 F.3d 1220, 1231 (11th Cir. 2018). Circuits have consistently given schools broad authority to punish students’ internet speech, regardless of its place of origin, so long as the school determines that the speech’s effects are or might be disruptive to the school. Disruptive is given a broad construction and left largely to the discretion of the school.⁶⁴See, e.g., Doninger, 527 F.3d at 54.

          1.     The Second Circuit

The major case for the Second Circuit is Doninger v. Niehoff.⁶⁵527 F.3d 41 (2d Cir. 2008). Doninger, like the precedent it relies on, holds that the “Tinker standard” applies, but it construes that standard as a substantial effects test.⁶⁶Id. at 50. Doninger relies heavily on Wisniewski ex rel. Wisniewski v. Bd. of Educ., 494 F.3d 34, 38–40 (2d Cir. 2007). See Doninger, 527 F.3d at 48–51. Wisniewski likewise construed the Tinker test as a substantial effects test only, without reference to the physical “at school” requirement, but did not explain the court’s rationale. See Wisniewski, 494 F.3d at 38. A more complete explanation is given in Doninger. See Doninger, 527 F.3d at 48–49. Doninger involved a student, Avery, who was a member of the Student Council.⁶⁷Doninger, 527 F.3d at 44. A school event called Jamfest suffered venue and scheduling issues.⁶⁸Id. Avery, along with four other Student Council members, sent a mass email to other students and the community, requesting that recipients contact the district superintendent to express support for Jamfest.⁶⁹Id. Both the superintendent and the principal were inundated with phone calls and emails, which annoyed them.⁷⁰Id. at 44–45. As a result, the principal spoke with Avery about the professionalism of sending out a mass email rather than discussing the issue with her or the superintendent directly.⁷¹Id. at 45.

Afterward, Avery posted on her personal, publicly accessible blog that Jamfest had been canceled altogether “due to douchebags in central office.”⁷²Id. She claimed that the principal had indicated that Jamfest was canceled.⁷³Doninger, 527 F.3d at 44–45. The principal denied ever having given such an indication, and the court agreed with the principal’s version of the facts. Her post included a letter for readers to “get an idea of what to write if you want to write something or call [the superintendent] to piss her off more. im [sic] down.”⁷⁴Id. at 45.

When this blog post was brought to the principal’s attention, she decided that Avery had “failed to display the civility and good citizenship” that class officers should demonstrate.⁷⁵Id. at 46. She told Avery to apologize to the superintendent and withdraw her candidacy for Senior Class Secretary.⁷⁶Id. Avery complied with the written apology but refused to withdraw her candidacy.⁷⁷Id. In response, the principal refused to endorse her nomination, effectively barring her from running.⁷⁸Id. The school also barred Avery from putting her name on the ballot or giving a campaign speech at the school assembly for elections.⁷⁹Doninger, 527 F.3d at 46.

The principal pointed primarily to the post’s “vulgar language” and the implied “encouragement of others to contact the central office ‘to piss [the superintendent] off more,’” as the reasons for Avery’s punishment.⁸⁰Id. (quoting Doninger v. Niehoff, 514 F. Supp. 2d 199, 208 (D. Conn. 2007)). The first reason the principal gave was that Avery failed to heed her counsel about how to resolve disagreements with the administration, but the other two reasons were about the language Avery used in her post. Avery was punished, at least in part, because the administrators found her language offensive. And while her behavior was not above reproach, both the district and appellate courts acknowledged that perhaps her punishment was too harsh for the offense.⁸¹See Doninger, 527 F.3d at 54 (“We . . . acknowledge her belief that in this case, ‘the punishment did not fit the crime.’” (quoting Doninger, 514 F. Supp. 2d at 202)). However, the Second Circuit left that to the school’s discretion.⁸²Id.

The court upheld Avery’s punishment because it dismissed the possibility that “territoriality,” or the physical “at school” nexus of Morse, could apply to internet speech.⁸³Id. at 48–49. In a previous decision, the Second Circuit had decided that “territoriality is not necessarily a useful concept in determining the limit of [school administrators’] authority.”⁸⁴Id. (quoting Thomas v. Bd. of Educ., 607 F.2d 1043, 1058 n.13 (2d Cir. 1979)). In Doninger, the court proceeded to remark that this observation was “even more apt today” given students’ engagement with the “school community” through the internet and e-communication.⁸⁵Id. at 49.

Dismissing even the possibility of applying the “at school” requirement physically, the court was left only to decide whether Avery’s speech might “foreseeably create a risk of substantial disruption within the school environment.”⁸⁶Id. at 48 (quoting Wisniewski ex rel. Wisniewski v. Bd. of Educ., 494 F.3d 34, 40 (2d Cir. 2007)). Because the post was “potentially disruptive of efforts to resolve the ongoing controversy” between the administration and students, the administrators had to deal with the communications traffic, and students were upset about the supposed cancellation of Jamfest, the court found that the post had been substantially disruptive.⁸⁷Doninger, 527 F.3d at 50–51. The court also upheld the lower court’s finding that Avery’s post was “purposefully designed by Avery to come onto the campus,” though designed off campus.⁸⁸Id. at 50. Thus, it was “reasonably foreseeable” that the post would have disruptive effects on school relations.⁸⁹Id.

          2.     Problems with the Second Circuit’s Approach

On the surface, the Doninger decision makes sense. Avery’s behavior was unprofessional, and it is understandable that the school administration would perceive her to be unfit to remain a school officer for her senior year. However, the reasoning of the decision is unsound and makes ex ante line drawing impossible.

First, the decision dismisses a critical component of the Tinker standard with a bare nod to “usefulness.”⁹⁰Id. at 48–49. The point of Tinker, Fraser, and Morse is that students have certain constitutional rights that can only be restricted within the school environment because of the unique concerns of schools.⁹¹See discussion of Tinker, Fraser, Hazelwood, and Morse, supra Part I. By insisting that students do not shed their rights “at the schoolhouse gate” and relying on an “at school” link in its decisions, the Supreme Court has indicated that a physical nexus of the speech to the school is necessary for school authority to operate.⁹²See Morse v. Frederick, 551 U.S. 393, 396–97, 401 (2007); Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 271 (1988); Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 506 (1969). The Supreme Court has not indicated in any way that this physical connection is obsolete, even as late as 2007.⁹³See Morse, 551 U.S. at 393. If the student in Morse had displayed his banner elsewhere, not at a school-supervised event and not on school grounds, the Court likely would have ruled that the school did not have the authority to suspend him, given the logic of the opinion.⁹⁴See id. at 396–97, 401; see also discussion supra Part I. The Second Circuit would disagree with my projection because its line of precedents has treated the territoriality distinction as much less significant than the substantial disruption part of the Tinker standard. Its position is that, because the Supreme Court has not explicitly stated whether off-campus speech is regulable by schools, one cannot say from Supreme Court precedent whether off-campus speech is outside school regulation. See Doninger v. Niehoff, 642 F.3d 334, 338 (2d Cir. 2011).

Second, the Second Circuit does not offer reasons why the territorial distinction should be abandoned. The court only says that students engage in online communication both on campus and off campus; therefore, the distinction is meaningless.⁹⁵See Doninger, 527 F.3d at 349. However, the fact that the internet can be accessed from anywhere does not inescapably lead to the conclusion that the school’s power to regulate must be coextensive with the internet’s reach. As the Third Circuit has noted, there is something concerning about allowing the school’s power to reach into the bedrooms of its students and regulate their internet speech.⁹⁶See Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 216 (3d Cir. 2011) (en banc), discussed infra Part II.C. Schools can and do regulate internet use on school grounds.⁹⁷See Alyson Klein, Schools Say No to Cellphones in Class. But Is It a Smart Move?, Educ. Wk. (Sept. 6, 2019), https://perma.cc/AK6J-MYMN; see also J.S. ex rel. Snyder v. Blue Mt. Sch. Dist. (Snyder), 650 F.3d 915, 921 (3d Cir. 2011) (en banc) (“The School District’s computers block access to MySpace, so no Blue Mountain student was ever able to view the profile from school.”). But to extend the school’s authority beyond that scope should require a concrete reason.

The Second Circuit in Doninger rules as if the second part of the standard, the substantial disruption to school relationships and operations, provides all the reason necessary to extend this authority.⁹⁸See Doninger, 527 F.3d at 48, 50–51. However, this reasoning makes ex ante line drawing impossible. If there is no territorial limit on what may be considered under the substantial disruption component, then there is no way for either the schools, the students, or the courts to determine the boundary between regulable and non-regulable speech before a situation arises. Everything must be determined on a case-by-case basis, first by the schools, then second guessed by the courts. Under the Second Circuit’s reasoning, any internet speech by a student about their school, made anywhere in the world, could have effects on school relationships with other students and with teachers or principals, and conceivably could fall under the school’s disciplinary powers. Given that the Second Circuit also gives the schools wide latitude in exercising their power,⁹⁹See id. at 54. the students might have little forewarning of what speech might get them in trouble.¹⁰⁰Note some types of speech might be covered by school policies or state and local laws (such as threats or harassment), and students would be on notice that their behavior is disfavored or wrong in those cases. See discussion infra Part III.B. However, this still does not eliminate the problems regarding the scope of authority. This only heightens the sense of injustice students might feel at being punished for their speech¹⁰¹Cf. Doninger, 527 F.3d at 54 (regarding Avery’s belief that her punishment was too harsh for the offense). and could raise concerns about arbitrariness.¹⁰²Cf. Kolender v. Lawson, 461 U.S. 352, 361 (1983) (holding that the relevant provision “encourages arbitrary enforcement by failing to describe with sufficient particularity what a suspect must do in order to satisfy the statute”). A further discussion of arbitrariness is beyond the scope of this Comment.

          3.     The Fourth, Eighth, Ninth, and Eleventh Circuits

The Second Circuit is not alone in its interpretation of students’ internet free speech rights. The Fourth, Eighth, Ninth, and Eleventh Circuits have also extended the school’s authority beyond the school’s domain. Their reasoning shares much the same issues as the Second Circuit’s reasoning.

The Fourth Circuit adopted the Second Circuit’s test of whether it was “reasonably foreseeable” for the off-campus internet speech to be brought to campus or to substantially disrupt relationships and operations on campus, relying significantly on the Doninger decision.¹⁰³See Kowalski v. Berkeley Cty. Sch., 652 F.3d 565, 571, 574 (4th Cir. 2011). The Fourth Circuit interpreted the Tinker standard as saying that the rights of students are not coextensive with adult’s rights, rather than that the rights of students in public school are not automatically coextensive with adults’ rights in the greater society.¹⁰⁴Compare id. at 571, with Bethel Sch. Dist. No. 403 v. Fraser, 484 U.S. 675, 682 (1986) (“[W]e reaffirmed that the constitutional rights of students in public school are not automatically coextensive with the rights of adults in other settings.”). Note, the Kowalski opinion incorrectly attributes this phrasing to Tinker itself, but the language is actually from Fraser. Justice Stewart’s concurrence in Tinker criticizes the Court’s “uncritical assumption that, school discipline aside, the First Amendment rights of children are co-extensive with those of adults.” Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 515 (1969) (Stewart, J. concurring). But Kowlaski refers to the main opinion of Tinker, not a concurrence, which is incorrect. The court read the modifying clause “in public school” more as establishing a legal status (public school students as opposed to non-students or adults) rather than specifying a location (in public school as opposed to at home or on a public sidewalk) that set a limited territorial sphere for the school’s power to curtail students’ freedom of speech.¹⁰⁵See Kowalski, 652 F.3d at 571. With this interpretation, the court was not overly particular about the physical location of the student’s internet activities. While acknowledging that there “is surely a limit” to a school’s interests in regulating off-campus speech, the court did not define that limit, finding that the substantial disruption element provided enough of a nexus to the school.¹⁰⁶Id. at 573. Kowalski had created a MySpace group webpage that mostly targeted another student, Shay N., with false and malicious dialogue. Kowalski created the page from her home computer while at home, and she shared it with several of her classmates (though not Shay N.). However, Shay’s parents quickly found out about the page, and Shay understandably felt harassed and humiliated. The school gave Kowalski a ten-day out-of-school suspension and a ninety-day “social suspension,” concluding she had created a “hate website,” violating the school policy against “harassment, bullying, and intimidation.” Id. at 568–69. The court upheld her punishment, ruling that a substantial disruption was reasonably foreseeable because other students accessed the MySpace page at school from their electronic devices and one student accessed it from a school computer. See id. at 569, 574.

The Eighth Circuit, also relying on the Second Circuit’s decision in Doninger, likewise interpreted Tinker as allowing regulation of speech, regardless of its location, if it satisfied the substantial disruption test.¹⁰⁷D.J.M. v. Hannibal Sch. Dist., 647 F.3d 754, 765 (8th Cir. 2011); see also S.J.W. v. Lee’s Summit R-7 Sch. Dist., 696 F.3d 771, 776–77 (8th Cir. 2012). The Eighth Circuit goes a little further than the Fourth Circuit by not expressly acknowledging a limit on the off-campus scope of the school’s disciplinary authority. Rather, the Eighth Circuit strongly implies that the physical location does not matter, interpreting the phrase “conduct by the student, in class or out of it, which for any reason . . . materially disrupts classwork or involves substantial disorder”¹⁰⁸Tinker, 393 U.S. at 513 (partially quoted by D.J.M., 647 F.3d at 765). as allowing regulation of any disruptive student conduct, regardless of where it occurred.¹⁰⁹D.J.M., 647 F.3d at 765. The speech at issue in D.J.M. v. Hannibal Public School District¹¹⁰647 F.3d 754 (8th Cir. 2011). was considered a true threat, and police took the student into custody fearing that he would put his words into action.¹¹¹Id. at 759, 762. Possibly the nature of the speech at issue might have affected how stringently the judges framed their test.

The Ninth Circuit has left open the question of whether school discipline reaches to all off-campus speech.¹¹²Wynar v. Douglas Sch. Dist., 728 F.3d 1062, 1069 (9th Cir. 2013). However, the trend in the Ninth Circuit indicates that courts in that circuit will continue broadening the school’s disciplinary reach. Originally, the Ninth Circuit stated that it was “reluctant” to create a “one-size fits all approach” to determining how much a school could regulate student off-campus speech.¹¹³Id. In Wynar v. Douglas School District,¹¹⁴728 F.3d 1062 (9th Cir. 2013). much as in the Eighth Circuit D.J.M. case, the student’s internet speech was classified as a true threat, expressing a desire to commit violence toward other students.¹¹⁵Id. at 1065–66. And as in D.J.M., police took the student in Wynar into custody based on those threats.¹¹⁶Id. at 1066. Perhaps the threatening nature of the speech at issue influenced how broadly the court considered the school’s disciplinary reach. For a while, the court was careful to leave open the possibility that the physical origin of the speech could play a role in deciding whether a school had authority to punish it, but the court ultimately decided Wynar solely based on the substantial disruption standard used by the other circuits.¹¹⁷Id. at 1068–70.

Since then, the Ninth Circuit has extended school disciplinary authority even to non-internet speech occurring off campus, so long as it might cause substantial problems for other students at school.¹¹⁸See McNeil v. Sherwood Sch. Dist. 88J, 918 F.3d 700 (9th Cir. 2019); C.R. v. Eugene Sch. Dist. 4J, 835 F.3d 1142 (9th Cir. 2016). Specifically, the court has upheld school discipline for a student who wrote a hit list of students in his journal (and who was also apprehended by police)¹¹⁹McNeil, 918 F.3d at 704–05. and for a student who verbally bullied other students in a park nearby the school.¹²⁰C.R., 835 F.3d at 1146–47. The extension of school authority to oral or physically written speech, which is easily defined territorially, suggests that the Ninth Circuit may continue to extend school disciplinary authority to nonthreatening internet speech should the case arise.

Finally, the Eleventh Circuit has applied Tinker to internet speech, but only in the context of college students. As regulating the speech of college students implicates other concerns not necessarily present in regulating the speech of middle school or high school students, this Comment generally avoids consideration of college cases. However, the Eleventh Circuit’s decision in Doe v. Valencia College¹²¹903 F.3d 1220 (11th Cir. 2018). strongly indicates how the court might rule if presented with an internet speech case for a younger student. In Doe, the court decided that “[t]he pervasive and omnipresent nature of the Internet has obfuscated the on-campus/off-campus distinction . . . making any effort to trace First Amendment boundaries along the physical boundaries of a school campus a recipe for serious problems in our public schools.”¹²²Id. at 1231 (quoting Bell v. Itawamba Cty. Sch. Bd., 799 F.3d 379, 391, 395–96 (5th Cir. 2015) (en banc)). This Comment will discuss the Fifth Circuit decisions separately since that circuit presents an oddity. The court’s rejection of the territorial distinction is similar to the other circuit courts of appeals’ reasoning: the substantial disruption part of the test is elevated, and the physical distinction is obliterated by the internet’s ubiquitous nature.¹²³Id. While the court acknowledges that the school’s power may not extend to all off-campus speech, it relies on the substantial disruption test to determine the boundary.¹²⁴Id.

B.     Group 2: Student Intent Governs the Territorial “At School” Nexus

The Fifth Circuit is still developing a standard for determining what off-campus speech schools may regulate. On one hand, the Fifth Circuit acknowledges the importance of the on-campus versus off-campus distinction for speech and rightly interprets Tinker and Fraser as requiring a close connection to the school for student speech to be restricted.¹²⁵See Bell, 799 F.3d at 389–90, 392–93. On the other hand, the Fifth Circuit believes that technological marvels such as “the Internet, cellphones, smartphones, and digital social media” make physical delineations “a recipe for serious problems” in defining the boundaries of school authority over speech.¹²⁶Id. at 392, 396 (quoting Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 220–21 (3d Cir. 2011) (en banc) (Jordan, J. concurring)).

Based on the facts of Bell v. Itawamba County School Board,¹²⁷799 F.3d 379 (5th Cir. 2015) (en banc). the Fifth Circuit decided that “when a student intentionally directs at the school community speech reasonably understood by school officials to threaten, harass, and intimidate a teacher,” the substantial disruption component is sufficient to allow school discipline, even if “such speech originated, and was disseminated, off-campus without the use of school resources.”¹²⁸Id. at 395–97. Bell posted a recording of a rap he created which contained threatening and obscene language toward two teachers. The lyrics referred to possible gun violence and falsely attributed obscene acts to the teachers. He posted the recording to his Facebook profile and then to YouTube, both of which were publicly accessible platforms. But he created the postings and the recording off school grounds. The court upheld his seven-day suspension and removal to the county’s alternative school. Id. at 383–86. In essence, this is a three-element test: (1) the student must intentionally direct his speech toward the school, intending that his speech reach the school community; (2) the speech must be reasonably understood as threatening violence; and (3) it must be reasonably foreseeable that the speech might cause substantial disruption.¹²⁹See id. In other cases where the students’ speech did not threaten violence, the court has either discussed only the first element of student intent (and found that the facts failed to satisfy this first element because the student did not intend for his speech to reach campus),¹³⁰See Porter v. Ascension Par. Sch. Bd., 393 F.3d 608, 615, 617 (5th Cir. 2004). or it has declined to reach the First Amendment issue.¹³¹See Jackson v. Ladner, 626 F. App’x 80, 89 (5th Cir. 2015). This case was decided one month after the en banc decision in Bell. The judges declined to reach the First Amendment issue because, regardless of how they might rule, they would find for the defendants on qualified immunity. Id.

The problem with the Fifth Circuit’s formulation of the on-campus versus off-campus distinction is that it relies on student intent to “reach the school community.”¹³²Bell, 799 F.3d at 396. Under this formulation, the Tinker, Fraser, and Morse emphasis on the speech’s physical nexus is transmuted to a relational nexus between the speaker and an undefined group. This formulation implies that the territorial constraint of the school’s power is eliminated. Conceivably, a school could punish a student for speech made in his home or sent in private messages to a friend, so long as those messages might affect in-school relationships. This formulation also implies that the court does not define the scope of “school community.” How many of the other students or teachers must a student reach before he reaches the “school community”? If he sends a tweet to a friend, who also happens to attend his school, is that sufficient? Or must a student keep all potentially disruptive communications private, as the student intended to in Porter v. Ascension Parish School Board?¹³³393 F.3d 608, 611–12 (5th Cir. 2004) (en banc). The student made a violent and disparaging drawing of his school while in his home and showed it to his mother, brother, and a friend who was living with them. He never intended the drawing to leave his house, nor did he show it to anyone outside of his household. The court ruled that his expression remained “off-campus” because he never intended for it to reach his school; it was accidentally taken there by his brother without his knowledge or consent. The answers to these questions remain unclear. The Fifth Circuit is still developing this standard, but currently it does not provide a sufficient rule for schools to follow.

C.     Group 3: Territorial Nexus Is Important, but Undecided on How Important Compared to the Substantial Disruption Prong

The Third Circuit’s jurisprudence remains muddled on whether schools can punish off-campus speech and to what extent. In two en banc decisions issued on the same day in 2011, the Third Circuit decided that the students’ punishments were unconstitutional because there was no substantial disruption at the schools and because all the internet content had been created off campus.¹³⁴Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 219 (3d Cir. 2011) (en banc); J.S. ex rel. Snyder v. Blue Mt. Sch. Dist., 650 F.3d 915, 933 (3d Cir. 2011) (en banc). In Layshock, the student accessed the website at times during class, but the school did not argue that the site became on-campus speech by the physical access point, so the court did not consider this argument. Layshock, 650 F.3d at 209. In Snyder, the schools blocked access to MySpace from school internet, so none of the students could access the site at school. Snyder, 650 F.3d at 921. Both decisions rely on the substantial disruption test, like other circuits. However, both cases emphasized that some physical nexus to the school is a necessary element for the scope of school disciplinary power, though they declined to delineate the precise boundaries of that power.¹³⁵Layshock, 650 F.3d at 216, 219; Snyder, 650 F.3d at 933.

In J.S. ex rel. Snyder v. Blue Mountain School District,¹³⁶650 F.3d 915 (3d Cir. 2011) (en banc). the Third Circuit interpreted Fraser as an exception to Tinker: even if the substantial disruption standard of Tinker was not met, schools could censor lewd speech from its students.¹³⁷Snyder, 650 F.3d at 932–33. The court ruled that applying this interpretation of Fraser to off-campus speech meant adopting a rule “that allows school officials to punish any speech by a student that takes place anywhere, at any time, as long as it is about the school or a school official, is brought to the attention of a school official, and is deemed ‘offensive’ by the prevailing authority.”¹³⁸Id. at 933. Snyder’s standard appears to be a two-step process: first, determine if the speech caused a substantial disruption; then, if it did not, consider the physical origin of the speech.¹³⁹See id. at 930–33. This is the opposite order of consideration from the proposed solution in this Comment. The case does not definitively answer whether a school could punish a student’s off-campus internet speech if it did cause a substantial disruption. The Third Circuit considered the Second Circuit’s Doninger decision and two other courts of appeals’ decisions relating to non-internet speech in a way that might indicate possible agreement, but the Third Circuit did not expressly adopt those approaches.¹⁴⁰Id. at 930–31.

In Layshock v. Hermitage School District,¹⁴¹650 F.3d 205 (3d Cir. 2011) (en banc). the school argued that the student’s speech originated on campus because he had taken a picture of the principal (whom the student’s internet site disparaged) from the school’s website.¹⁴²Id. at 214–15. The Third Circuit ruled that this was simply too much of a stretch.¹⁴³Id. at 216. While “Tinker’s ‘schoolhouse gate’ is not constructed solely of the bricks and mortar surrounding the school yard,” the court recognized that “the concept of the ‘school yard’ is not without boundaries,” even territorial ones.¹⁴⁴Id. The court refused to create precedent allowing the school to extend its reach into students’ homes to censor them:

It would be an unseemly and dangerous precedent to allow the state, in the guise of school authorities, to reach into a child’s home and control his/her actions there to the same extent that it can control that child when he/she participates in school sponsored activities. Allowing the District to punish [this student] for conduct he engaged in while at his grandmother’s house using his grandmother’s computer would create just such a precedent . . . .¹⁴⁵Id.

The court also implicitly rejected the Fifth Circuit’s intent nexus, stating that the student’s “aim[ing]” the online profile at the principal was insufficient to constitute on-campus speech.¹⁴⁶Id. at 216–17. The student was careful to keep his internet activities away from campus, and merely aiming his offensive remarks at a school official did not obliterate the on-campus versus off-campus distinction.¹⁴⁷Layshock, 650 F.3d at 216–17. But while Layshock placed great importance on the on-campus versus off-campus distinction, the decision relied heavily on the finding that the speech had not caused substantial disruption at school.¹⁴⁸Id. at 214. Thus, it too left open the question of whether the territorial distinction would be relevant if the speech had caused a substantial disruption.¹⁴⁹Id. at 219 (“We need not now define the precise parameters of when the arm of authority can reach beyond the schoolhouse gate because, as we noted earlier, the district court found that Justin’s conduct did not disrupt the school, and the District does not appeal that finding.”).

The confusion created by Snyder and Layshock is evident in subsequent cases at the district court level. One such case discussed both en banc decisions in detail, pointing out that both decisions “failed to resolve the question of whether [the substantial disruption test from] Tinker ‘can be applicable to off-campus speech.’”¹⁵⁰R.L. ex rel. Lordan v. Cent. York Sch. Dist. (Lordan), 183 F. Supp. 3d 625, 633–34 (M.D. Pa. 2016) (quoting Layshock, 650 F.3d at 220 (Jordan, J. concurring)). Left without any guidance on the question, the district court proceeded to consider the precedents from the Second, Eighth, and Ninth Circuits, and it decided to follow them in applying the substantial disruption test to off-campus internet speech.¹⁵¹Id. at 638–40. In spite of lower court confusion, the Third Circuit has made no further rulings on the issue to date.¹⁵²There is no subsequent appellate history for Lordan, so the Third Circuit did not express its opinion in that case.

III.     Analysis

There are two major problems with how the circuit courts of appeals have handled student internet speech cases. First, the circuits have moved away from the “at school” element of Tinker and its subsequent cases.¹⁵³See, e.g., Doninger v. Niehoff, 527 F.3d 41, 48–49 (2d Cir. 2008). Their rationale is that the internet is ubiquitous, accessible from any place; therefore, physically distinguishing between “at school” and away from school is meaningless.¹⁵⁴See, e.g., id. Second, the circuits have extended the scope of the effects test part of Tinker.¹⁵⁵See, e.g., id. at 53. Tinker says nothing about a school’s ability to regulate off-campus speech; it only provides standards for regulating on-campus speech.¹⁵⁶See Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 505–14 (1969); see also Clay Calvert, Off-Campus Speech, On-Campus Punishment: Censorship of the Emerging Internet Underground, 7 B.U. J. Sci. & Tech. L. 243, 264–65 (2001) (“The United States Supreme Court, however, has never articulated standards regarding how much authority a public school may assert over off-campus student expression.”). But instead of determining the on-campus or off-campus nature of the speech before considering the amount of disruption caused, the courts look primarily to the disruptive (or potential disruptive) effect of the speech and decide that speech that is disruptive is on-campus speech per se.¹⁵⁷See, e.g., Doninger, 527 F.3d at 53. As discussed supra Part II.C., the Third Circuit is undecided on this issue, but a future decision following the other courts in this analysis would remain consistent with its formulation in Snyder.

The courts’ reliance on the substantial disruption factor, in the absence of further guidance from the Supreme Court, has resulted in the courts’ deciding whether speech is regulable on a case-by-case basis after a school has decided whether to take action. Essentially, the courts have told the schools to go ahead and regulate speech, then ask for forgiveness later. Some of the courts of appeals are beginning to realize that the schools’ power has grown too extensive.¹⁵⁸See, e.g., Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 216 (3d Cir. 2011) (en banc). However, even these courts have not drawn a satisfactory line based solely on the speech’s effects at school.

A.     The Solution

Because the current standards have proved unworkable, this Comment proposes a different approach to dealing with internet speech. First, the on-campus versus off-campus distinction based on a territorial nexus should be revived. Second, the schools should focus less on punishing the internet speech—that is, the actual words posted on an internet forum—and more on punishing on-campus conduct which causes disturbances or harm to others.

Almost uniformly, the courts of appeals have brushed aside the very idea that a territorial nexus can be applied to internet speech, often with little reasoning to support this framework-altering decision.¹⁵⁹See discussion supra Part II.A-B. Their reasoning is based on the fact that the internet is accessible everywhere and not tangible; therefore, it is incapable of geographical restriction.¹⁶⁰See discussion supra Part II.A-B. However, this formulation misses the point of the territorial component to the on-campus versus off-campus distinction. The territorial nexus is not about the reach of the speech or the medium conveying the speech; it is about the reach of the school’s authority.¹⁶¹See Tinker, 393 U.S. at 511. The fact that the internet is nearly ubiquitous or that school relationships may extend beyond the school context does not compel the conclusion that the school’s disciplinary authority must be similarly coextensive.

Other scholars have also suggested reviving the on-campus versus off-campus distinction for internet speech. Professor Clay Calvert suggests a five-part solution to the courts’ conundrums over internet speech; two of those parts involve a revived geographical component to the analysis.¹⁶²Calvert, supra note 156, at 264–67. His solution connects the territorial nexus to both the place of origin of the speech (i.e., whether the student was on school property while speaking online) and the place of download or access (i.e., whether the student ever downloaded or accessed their speech on school machines or on school grounds).¹⁶³Id.

Professor Barry McDonald sharply criticizes analyzing cyberspeech cases “without regard to where the speech occurred,” saying the courts engaging in such analysis are “missing the boat” and often applying more restrictive rules where they should apply less restrictive ones.¹⁶⁴Barry P. McDonald, Regulating Student Cyberspeech, 77 Mo. L. Rev. 727, 750–51 (2012). And while Professor Susan Bendlin acknowledges that the location of the internet speaker may not be dispositive of the physical nexus issue, she concludes that “when the speech itself does not occur during a school-supervised activity, it is considered to be off-campus, and the school’s disciplinary arm should not reach it. Tinker should not be extended to cover the creation of off-campus social networking messages.”¹⁶⁵Susan S. Bendlin, Far from the Classroom, the Cafeteria, and the Playing Field: Why Should the School’s Disciplinary Arm Reach Speech Made in a Student’s Bedroom, 48 Willamette L. Rev. 195, 217–18, 222 (2011).

The rationale behind limiting the school’s authority to the school’s environs is simple: students are more than just students. Each of these young people has a life outside the classroom. Outside the school, they should be primarily considered citizens, not students.¹⁶⁶See Calvert, supra note 156, at 265, 271. The courts of appeals often seem to have forgotten this principle in their decisions, analyzing minors only in their roles as students even when they engage in non-scholastic activities.¹⁶⁷See, e.g., Kowlaski v. Berkeley Cty. Sch., 652 F.3d 565, 567–69 (4th Cir. 2011); J.S. ex rel. Snyder v. Blue Mt. Sch. Dist., 650 F.3d 915, 929–31 (3d Cir. 2011) (en banc); but see Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 216 (3d Cir. 2011) (en banc). Extending the school’s authority beyond the school encroaches on the liberty that minors enjoy as persons under the protection of the Constitution.¹⁶⁸See Calvert, supra note 156, at 271.

The second component of the proposed solution is that schools should have no ability to punish internet speech itself, except for content posted on a site belonging to or administered by the school.¹⁶⁹This exception is reasonable, particularly in light of the Supreme Court’s ruling in Hazelwood that schools have a particular interest in speech to which they ascribe their name. See Hazelwood Sch. Dist. v. Kuhlmeier, 484 U.S. 260, 270–73 (1988). Undoubtedly, schools should have the ability to punish on-campus conduct which disrupts the school, and they do. Under this solution, a school could punish a student for repeating internet speech to bully someone or for using phones or computer stations in a disruptive manner while at school. Thus, if a student badgered another through instant messaging while in school, the school would have the authority to confiscate his phone or give him detention. However, the schools should not gain unlimited authority over students’ internet activity simply because it might be accessed by someone at school.

Here, this Comment departs from Professor Calvert’s opinion that place of download or access should be a factor when analyzing the scope of the school’s authority.¹⁷⁰Calvert, supra note 156, at 265–67. Authorizing a school to punish internet speech that is merely downloaded or accessed at school gives a school too much power over what students can say online. To illustrate, consider a student, M, who is in study hall.¹⁷¹This hypothetical is invented and not based on any real case, but it is a plausible situation. Suppose M gets bored and, taking a short break from studying, quietly pulls out her phone and opens her Twitter app. Suppose also that the night before, while at her house, she saw a tweet from a friend B poking fun at a mutual acquaintance X, who goes to her school. She decided to reply with something mean about X. Now, during her study break at school, M sees and reads X’s reply. However, M’s comment from the night before is included in the comment chain she is now viewing on her phone.

Under the place of download or access test, the school could punish M for her tweet, since she accessed and downloaded it on her phone while at school. And under the courts’ current formulations of the substantial disruption test, if X or even third parties were disturbed by the tweets, the school would have a case for punishing M for her tweet. Yet this hardly seems fair. M tweeted her comment from the privacy of her home, and it just happened to come up again in her feed while at school. But because both Twitter and the interpersonal relationships between M, X, and B extend both within the school and outside the school, M’s school claims authority to punish M for her tweet.

M’s case is hypothetical, but it is similar to what has occurred in actual cases.¹⁷²Cf. Jackson v. Ladner, 626 F. App’x 80, 82–84 (5th Cir. 2015); Thomas v. Bd. of Educ., 607 F.2d 1043, 1045–46 (2d Cir. 1979); Beussink v. Woodland R-IV Sch. Dist., 30 F. Supp. 2d 1175, 1177–79 (E.D. Mo. 1998). For example, under the place of download or access test, Justin Layshock’s internet speech should have been considered on campus after all, since he accessed the MySpace page from a classroom computer and showed the page to other students at school.¹⁷³Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 209 (3d Cir. 2011) (en banc). The court did not consider this argument, since the school administration raised a different reason why Layshock’s speech should be on-campus speech (which was rejected). Id. at 215–19. However, the Third Circuit correctly concluded that Layshock’s speech was beyond the reach of the school’s authority,¹⁷⁴See id. at 216, 219. and even his accessing the page while at school should not negate that conclusion.

This is not to say that the school would be prevented from punishing M if she disruptively showed her Twitter exchange to X or her other classmates during class. On the contrary, the school can and should keep order, as settled in Tinker.¹⁷⁵Tinker v. Des Moines Indep. Cmty. Sch. Dist., 393 U.S. 503, 507, 512–13 (1969). However, the school would not punish M for her speech but for her disruptive conduct. And along the same lines, a school could punish a student for inciting or causing a substantial disruption, though the school could not censor off-campus speech.¹⁷⁶See Bendlin, supra note 165, at 222–24 (“The ‘effect’ is the result of the speech, but is not the speech itself. The school cannot suppress the message itself or discipline the speaker himself or herself unless that speech was made in a school-supervised setting.”).

B.     Testing the Solution

Critics may object to the proposed solution. First, they may ask, since the internet is ubiquitous, how is a school to keep order and prevent off-campus internet speech from causing disruptions in the school, if it cannot punish internet speech?

One answer is for schools to monitor more effectively what can be accessed through the school’s internet and computers. Darryn Cathryn Beckstrom provides examples of Internet Service Providers (“ISPs”) and software that allow parents to impose limits on their children’s online activities.¹⁷⁷Darryn Cathryn Beckstrom, State Legislation Mandating School Cyberbullying Policies and the Potential Threat to Students’ Free Speech Rights, 33 Vt. L. Rev. 283, 315–16 (2008). Schools could use such ISPs or software to deny entry to certain websites or, perhaps more practically, to allow entry only to certain websites.¹⁷⁸See id. at 316. If social media created too much distraction at school, the school could block access to those sites. Additionally, with some cases arising over students’ personal websites,¹⁷⁹See, e.g., Emmett v. Kent Sch. Dist. No. 415, 92 F. Supp. 1088, 1089 (W.D. Wash. 2000). a school could set up its computers to access only certain approved sites.

Another answer is for schools to restrict students’ access to their phones during school hours. Schools throughout the world are dealing in various ways with the challenges of smartphones, from complete bans¹⁸⁰For example, Forest Hills district near Grand Rapids, Michigan, and France for students under age fifteen. See Klein, supra note 97. to incorporating the phones into the lesson as teaching tools.¹⁸¹For example, Saint Mary’s Area school district in Pennsylvania. Id. Some schools have received resistance from parents regarding cellphone restrictions, as some parents want to be able to contact their children at any time, even during class.¹⁸²Id. However, cellphone restriction remains an option, and if a district is having difficulty keeping order in its schools, restricting phone access might help. Regardless, there are plenty of available options other than censoring or controlling student internet speech.

Next, critics may wonder how this solution addresses the phenomenon of cyberbullying, student-on-student online communications where one or more students harass another. Some scholars have advocated that schools have broad authority to punish cyberbullying because “the hurtful speech can be accessed anytime and anywhere, including at school.”¹⁸³See Jennifer Butwin, Note, Children Are Crying and Dying While the Supreme Court is Hiding: Why Public Schools Should Have Broad Authority to Regulate Off-Campus Bullying “Speech,” 87 Fordham L. Rev. 671, 699 (2018). And indeed, the Ninth Circuit gave schools broad authority to punish even in-person bullying that occurred off campus.¹⁸⁴See C.R. v. Eugene Sch. Dist. 4J, 835 F.3d 1142, 1153 (9th Cir. 2016); Butwin, supra note 183, at 699. Professor McDonald notes a particular problem with cyberbullying that is not present with in-person bullying—the bully and the bullied person may be in different places such that the online bullying may cross school borders.¹⁸⁵See McDonald, supra note 164, at 732. One person might be at home or at the mall, but the other might be sending or receiving the online messages at school.¹⁸⁶See id.

First, in dealing with cyberbullying, it is important to realize that no single solution will completely solve the problem. Even giving the schools unlimited power to punish bullying would not solve cyberbullying. Beckstrom notes that just as other scholars have suggested that “traditional bullying statutes are not particularly effective in preventing bullying,”¹⁸⁷See Beckstrom, supra note 177, at 313 (citing Kathleen Hart, Note, Sticks and Stones and Shotguns at School: The Ineffectiveness of Constitutional Antibullying Legislation as a Response to School Violence, 39 Ga. L. Rev. 1109, 1146 (2005)). there is reason to believe that cyberbullying statutes might be similarly ineffective.¹⁸⁸See id. But she provides several other solutions that, working together, could make a real impact.¹⁸⁹See id. at 313–14. Among these are increased parental awareness of cyberbullying and their child’s online habits, security controls for children’s browsing, and legal changes making egregious cyberbullying (often called cyberharassment or cyberstalking) subject to criminal charges and tort liability.¹⁹⁰Id at 314–20. All of these solutions are necessary to help combat cyberbullying, regardless of how the on-campus versus off-campus speech question is resolved.

Second, a school can still punish bullying, including cyberbullying, which occurs within its scope of authority. Bullying has a conduct element to it too, and as previously explained, a school has the authority to keep order and peace within its gates. The school could punish a student for bullying another, even if the student was using an online medium to bully, because the school would be punishing the student for their conduct, not their speech.

Of course, the school’s authority would be somewhat limited in Professor McDonald’s scenario. If the bully was at school and cyberbullied someone who was not at the school, restricting the bully’s access to certain sites and to cellphones might solve that problem without needing to address content issues. But in the reverse, if the bully was not at school and the bullied person was receiving the bullying while at school, it is difficult to see how the school could punish the bully, though restricting student access to online media during school hours might give the bullied person some relief from the bullying. However, even with the broad authority and discretion schools currently enjoy, dealing with this scenario is often beyond their ability.¹⁹¹Cf. Butwin, supra note 183, at 672–75. Curtailing the school’s power in the manner proposed would not make a significant difference in terms of effectiveness. This is why Beckstrom advocates not relying solely on school authority, but also on raising parental awareness, as a more effective way to prevent cyberbullying.¹⁹²Beckstrom, supra note 177, at 314.

The major advantage of the proposed solution is what is gained. Under the approach that the school must have broad authority to punish cyberbullying, one must find some way to protect the rights of the accused. When schools have broad authority to punish whatever they deem cyberbullying, they become both prosecutor and judge, and students’ rights may be in jeopardy.¹⁹³Calvert, supra note 156, at 276 (quoting Thomas v. Bd. of Educ., 607 F.2d 1043, 1051 (2d Cir. 1979)). None of the courts of appeals have articulated a sufficient safeguard at this time. However, restoring the territorial nexus to the on-campus versus off-campus distinction and limiting the school to punishing only expressive conduct occurring on campus provides a safeguard for student rights. Schools could not just punish whatever speech offended someone, pleading an effect on the school as justification. Since most speech could have some effect on school relationships, the schools currently can punish at their discretion and leave the courts to decide, after the fact, whether the effect was substantial enough to warrant school intervention.¹⁹⁴See discussion supra Part II; see also Doninger v. Niehoff, 527 F.3d 41, 54 (2d Cir. 2008). Under the proposed model, the school would first have to establish that the speech or expression was on campus. That determination is much more susceptible of an ex ante rule than the substantial disruption analysis.

For more serious forms of cyberbullying, which many jurisdictions have labeled cyberstalking and cyberharassment,¹⁹⁵See Joey L. Blanch & Wesley L. Hsu, An Introduction to Violent Crime on the Internet, 64 U.S. Att’ys’ Bull. 2, 5 (2016). victims have recourse in the legal system. In all fifty states, victims have criminal and civil recourse for stalking, and all states have passed cyberstalking or cyberharassment laws (or both).¹⁹⁶Cyberstalking, National White Collar Crime Center (NW3C), 1–2 (2015). Jennifer Butwin argues that legal remedies are less effective for bullying concerns because it can hurt victims more to face their bullies in court, and bullies might not be found liable for slander or libel.¹⁹⁷Butwin, supra note 183, at 695–96. However, her argument focuses on the kind of bullying where a bully might call another student hurtful names.¹⁹⁸Id. at 696. The more serious kinds of cyberbullying, which often are reported in the news and drive the creation of these laws, where the victims committed suicide at the insistence or influence of their bullies,¹⁹⁹Id. at 672–74. can be solved with criminal and civil penalties.

Butwin insists that schools are better equipped to handle bullying problems.²⁰⁰Id. at 693–97. But, as Butwin acknowledges, many cyberbullying incidents happen away from school.²⁰¹Id. at 672–75. School discipline cannot prevent such tragedies because schools simply cannot keep a leash on all their students all the time, especially when students are away from school. The only thing that probably could prevent cyberbullying suicides would have been more parental awareness of cyberbullying combined with fostering the students’ sense of self-worth.²⁰²A common theme in the bullying suicide stories that Butwin cites is that the parents were completely unaware of their children’s online lives and did not notice anything was wrong with their children before the students committed suicide. Id. at 672–74. This Comment does not at all blame the parents for these tragedies. The point of this Comment is that the only authorities who could have done something about the cyberbullying away from school were the parents of the children involved. Schools simply do not have the resources or ability to end cyberbullying, even if given the broad power that Butwin advocates. And where the bullying happens on campus, it would count as on-campus expressive conduct that a school could punish, under both our schemes.

Also, in a case of severe cyberharassment or cyberstalking, the bully is committing a crime. All fifty states have laws against this behavior.²⁰³Cyberstalking, supra note 196, at 2. If someone, even a teenager, engages in such malicious behavior as attempting to drive another student to self-harm or suicide, then they should receive the punishment their actions deserve. Courts, not schools, are best equipped to handle criminal behavior. It is surprising that Butwin urges that schools handle serious cyberbullying cases, since she also expresses concern that, if they escape punishment, bullies might learn there are no consequences for their actions.²⁰⁴Butwin, supra note 183, at 696. But even a suspension or expulsion from school would not be a true punishment for cyberharassment. That behavior is serious enough to merit serious punishment, which is the reason that all fifty states have passed laws criminalizing it.²⁰⁵Cyberstalking, supra note 196, at 2. Plus, school punishments could give a bully more opportunities to engage in cyberbullying away from school. Of the three major authority figures in a student’s life—parents, schools, and government—the schools seem the least able to handle severe cyberbullying precisely because of the internet’s ubiquitous nature. Schools can raise awareness among students and discourage cyberbullying,²⁰⁶See Bendlin, supra note 165, at 234. but the type of bullying Butwin describes would be beyond the school’s capabilities. Such persistent, pernicious behavior can only be handled by in-home authority or state laws.²⁰⁷Also note that Butwin’s opening case, that of Megan Meier, would have been completely beyond the abilities of a school because Megan’s bully was an adult woman masquerading as a teenage boy online. Butwin, supra note 183, at 673 n.11.

With regard to punishing milder forms of cyberbullying, Butwin is correct that criminal punishment for calling other students names online would be detrimental.²⁰⁸See id. at 695. However, it does not follow that schools should step in and punish where the courts cannot. Professor McDonald makes a good point when he states that any efforts to deal with cyberbullying should also consider that students are “immature minors” who at times will say “mean or hurtful things to other students.”²⁰⁹McDonald, supra note 164, at 757. Because students are immature, Professor McDonald argues that they should receive some protection from censorship to make mistakes and learn from them.²¹⁰Id. He notes that varying definitions of cyberbullying show that “drawing a line between cyberbullying and mere immature adolescent rudeness can be very difficult.”²¹¹Id. Students should have some “First Amendment ‘breathing space’” to learn from mistakes without government (or school) sanction.²¹²Id. Again, the case for increased parental involvement arises, since parents are the most effective teachers of acceptable social behavior; but this is not a solution that the law can impose.

This line between adolescent rudeness and harassment is also a problem with speech that Professor McDonald labels “cyberdissing”—student internet speech attacking school officials, personnel, or administration, instead of other students.²¹³Id. at 728. Many of the cyberspeech cases the courts of appeals have decided are “cyberdissing” cases.²¹⁴See Jackson v. Ladner, 626 F. App’x 80, 83 (5th Cir. 2015); Layshock v. Hermitage Sch. Dist., 650 F.3d 205, 209 (3d Cir. 2011) (en banc); J.S. ex rel. Snyder v. Blue Mt. Sch. Dist., 650 F.3d 915, 915 (3d Cir. 2011) (en banc); Doninger v. Niehoff, 527 F.3d 41, 46 (2d Cir. 2008). In these cases, the students were punished because the school authorities found the content offensive or inappropriate.²¹⁵See Jackson, 626 F. App’x at 83; Layshock, 650 F.3d at 209; Snyder, 650 F.3d at 922; Doninger, 527 F.3d at 46. Professor Calvert notes, however, that “what may be offensive speech to teachers may be amusing humor to students,”²¹⁶Calvert, supra note 156, at 284. as “one teacher’s threat is another student’s parody.”²¹⁷Id. Professor Calvert further explains that much of today’s student humor is rude, offensive, and “sick,” both online and in popular television programs.²¹⁸Id. at 283. Calvert mentions South Park as an illustrative television program. This Comment suggests Reddit, Tumblr, and meme culture generally, as illustrative internet speech. Punishing students for all speech that someone finds offensive heavily infringes on students’ speech rights. And there is something gravely wrong with allowing school officials to punish speech that criticizes them simply because they find it offensive. That form of “cyberdissing” could be akin to political speech, though perhaps about intra-scholastic “politics.” Or, at a lower level, it could simply be immature student humor about their life experiences.²¹⁹Cf. id. at 284. Either way, allowing schools to punish such speech smacks of abuse of power.

Critics may point to the case of Justin Swidler, the student plaintiff in J.S. v. Bethlehem Area School District,²²⁰J.S. v. Bethlehem Area Sch. Dist. (Swidler), 807 A.2d 847 (Pa. 2002). as a reason why schools should be able to punish “cyberdissing.” Swidler created a website on his home computer that denigrated one of his teachers and the principal with obscene language.²²¹Id. at 850–51. The site perfectly fits the definition of “off-campus”: it was created in Swidler’s free time, and he took pains to keep its existence hidden from the school.²²²Id. The website contained a disclaimer requiring the visitor to agree that they would not tell the school of the site’s existence or content, though the site was not password protected. Id. at 851. Someone disclosed the site’s contents to the teacher and the school administrators, and the school construed the contents as a threat to the teacher.²²³Id. at 852. The teacher was also deeply distressed by the contents, and she suffered from so much anxiety that she was unable to continue teaching for the rest of the year.²²⁴Id. The school suspended Swidler for three days, then extended that suspension to ten days.²²⁵Id. Shortly after, the school began expulsion proceedings against Swidler.²²⁶Swidler, 807 A.2d at 852.

Professor Calvert argues that even such egregious “cyberdissing” should not be punished by in-school proceedings. The message that society should teach young people is “[r]eal-world speech carries real-world consequences.”²²⁷Calvert, supra note 156, at 246. But, Professor Calvert argues, schools instead teach students that they can be punished twice for the same offense.²²⁸See id. at 246–48. He points out that “off-campus justice was pursued for off-campus speech.”²²⁹Id. at 248. The principal and the teacher who suffered so much distress sued Swidler for defamation and other torts.²³⁰Id. at 247–48. Additionally, the local police and the FBI investigated whether Swidler’s statements constituted a “true threat,” though both ultimately concluded that they did not because Swidler was not serious.²³¹Id. at 248. Swidler received real-world consequences for the harm he did to the school officials.

Yet he also faced punishment from the school because his statements had a “negative effect on other students’ perception of [the teacher and principal].”²³²Id. at 249 (quoting J.S. v. Bethlehem Area Sch. Dist., 757 A.2d 412, 421 (Pa. Commw. Ct. 2000)). Since the teacher and principal both recovered compensation through the legal system for the harm that Swidler did them,²³³The teacher recovered $450,000 in damages, and her husband recovered $50,000 for the loss of consortium claim, for a total of $500,000. Lauri Rice-Maue, Swidlers to Pay $500,000 for Web Site Jury Finds Fulmer Was Not Defamed But Her Privacy Was Invaded, Morning Call (Nov. 2, 2000), https://perma.cc/8MZ5-SRCX. The principal won $500,000 from his defamation lawsuit against the Swidlers. Precious Petty, Man in Legal Battle Over Website as Teen, Looks Back on Experience, Its Longterm Effect, Express-Times (Jan. 3, 2019), https://perma.cc/9LGC-49TF. it seems unnecessary and vindictive for the school to punish Swidler further by keeping him out of school. Professor Calvert contends that this practice of in-school punishment for “cyberdissing” based on possible effects at school was “tantamount to a heckler’s veto,” “trampl[ing]” the speaker’s rights over concern about the audience’s reaction.²³⁴Calvert, supra note 156, at 249.

This practice raises concerns because of its wide application. Swidler may be a very unsympathetic plaintiff; after all, his speech was extremely offensive and crude. However, schools have used the same justifications to punish students for such innocuous speech as criticizing their football team.²³⁵Watch What You Tweet: Schools, Censorship, and Social Media, Nat’l Coal. Against Censorship (Jan. 4, 2020), https://perma.cc/GCP3-K4TC (noting that the senior class president at Heights High School in Kansas was suspended for comparing the football team to Wichita State University’s team, which did not exist). Schools should not have the power to use their disciplinary authority as a heckler’s veto or as retaliation for being offended. Clarifying that the school’s power does not extend to censoring internet speech would better protect students’ free speech rights.

Finally, the last category of cyberspeech discussed in the courts of appeals cases, students speaking of violence against their schools, can also be better handled by reviving the territorial nexus. Once again, police are better at determining whether speech constitutes a real threat of violence or merely an opportunity for nonserious venting.²³⁶See, e.g., Wisniewski ex rel. Wisniewski v. Bd. of Educ., 494 F.3d 34, 36 (2d Cir. 2007) (noting that police determined that the offending icon was a joke and was not a threat, while the school determined otherwise); see also D.J.M. v. Hannibal Pub. Sch. Dist., 647 F.3d 754, 759 (8th Cir. 2011) (noting how local police took the student into custody and put him in juvenile detention and a psychiatric facility, to prevent his carrying out various threats); J.S. v. Bethlehem Area Sch. Dist., 807 A.2d 847, 852–53 (Pa. 2002) (noting that local police and FBI determined that speech was nonserious, while the school construed it as a threat). Where there is a real threat to the school, the police often take the student into custody, and the criminal justice system prevents any tragedies to the school.²³⁷See, e.g., D.J.M., 647 F.3d at 759; see also Wynar v. Douglas Cty. Sch. Dist., 728 F.3d 1062, 1066 (9th Cir. 2013). Schools do play a role in this apprehension; once the threats are brought to the school’s attention, the administrators are often the ones who alert the police.²³⁸See, e.g., D.J.M., 647 F.3d at 759; see also Wynar, 728 F.3d at 1066. But there is no need for additional punishment from the school for students who make true threats.

Indeed, for cases where the students are only venting frustration, school discipline may only exacerbate the situation. Professor Calvert argues that students need a method to vent their school frustrations that does not end in violence.²³⁹Calvert, supra note 156, at 282. One such “safety valve” is internet speech²⁴⁰Id.—student blogs, websites, webpages, and forums, which for this generation have become modern pen-and-paper notes and diaries. One fear is that the schools’ readiness to bring the full force of its disciplinary authority on any speech that hints at violence may actually result in more violence. If students find the avenue of speech closed to them, they may find other, more destructive ways of venting their emotions. As Professor Calvert notes, “[w]e should be thankful that [Swidler] was using speech and not a gun to express his emotions.”²⁴¹Id. Police have proved to be better than schools at finding the line between true threats and venting. In the four example cases cited above,²⁴²See supra notes 236–37 and accompanying text. the police properly distinguished between true threats that required intervention and false alarms, but the schools cried “true threat” every time.²⁴³See supra notes 236–37 and accompanying text. Barring schools from punishing this speech and leaving such prosecution to the authorities is a more reliable method of both protecting the schools and protecting students’ free speech rights.

Conclusion

Students have free speech rights, but the current scheme offers them little protection. The various tests articulated by the courts of appeals leave far too much power and discretion in the hands of school administrators. The courts have not given schools a clear rule to follow, so the schools punish whatever they think merits such action. Often, this is whatever a school administrator finds offensive, even if a student intended their speech as parody or humor. That is not freedom, and sometimes it leads to a double punishment for a student—one given through legal action and another heaped on by the school.

The courts have failed to articulate a clear rule because they assume that the internet’s intangible nature makes territorial line drawing impossible or absurd. However, this assumption is based on faulty reasoning which conflates the physical scope of school authority with the scope of the speech at issue. Tinker’s articulation of a territorial standard is meant to define the scope of the school’s authority, regardless of the scope of the students’ speech.

A better approach is to return to Tinker’s territorial model of regulation. School disciplinary authority should only reach on-campus speech and expressive conduct. Schools should be prohibited from punishing students for their internet speech, excepting only speech posted on school-operated or school-sponsored sites. To maintain order, schools can block access to certain sites from their school internet and computers, and they can police official school sites. Schools also can control smartphone use during school hours or events, and schools can punish students for causing or participating in substantially disruptive behavior on campus. But schools should not be allowed any further to regulate students’ internet speech or punish them for it.

Ultimately, the Supreme Court needs to define the scope of Tinker in its application to off-campus speech. The confusion among the lower courts will only continue if the Supreme Court does not act. For the sake of students’ First Amendment rights, the Court should refuse to jettison the territorial restriction from Tinker. Instead, the Court should re-emphasize the territorial nexus’s importance as the first step of analysis, for it is the most effective check on abuse of the schools’ disciplinary power.

Society also needs to stop leaving everything to the schools. As noted, particularly in cases of cyberbullying and cyberspeech hinting at violence, parental involvement is key to preventing tragedies and maintaining order. Schools should not be expected to reach into the home, and they are ineffective at preventing cyberbullying originating at home. Since internet communication is indeed ubiquitous, parents and students alike should become more aware of its dangers and actively seek to prevent tragedies like cyberbullying suicides. Extending the reach of school authority is not nearly as effective as proactive home authority, and such extension has been disastrous for students’ free speech rights. It is time to put limits on the school’s authority outside the school and stop using the ubiquitous nature of the internet as an excuse for punishing students’ internet speech.