Automated Student Monitoring and Equal Access to Public Education

Introduction

Public school districts are, with growing frequency, adopting surveillance software that monitors student expression on school networks and school-issued devices. 1 See Barbara Fedders, The Constant and Expanding Classroom: Surveillance in K-12 Public Schools, 97 N.C. L. Rev. 1673, 1674-75 (2019); Claire Bryan & Sharon Lurye, Schools Use AI to Monitor Kids, Hoping to Prevent Violence. Our Investigation Found Security Risks, AP News (Mar. 12, 2025, 4:00 AM PDT), https://perma.cc/Y3XV-TXHU; Kéah Sharma, Public Schools, Private Eyes: How EdTech Monitoring Is Reshaping Public Schools, New Am. (Aug. 25, 2025), https://perma.cc/PXN5-T3FN; Kirsten Kendrick, WA School Districts Monitoring Thousands of Students 24/7, KNKX Pub. Radio (Apr. 18, 2025, 5:00 AM PDT), https://perma.cc/Q243-UEBV. In an era of persistent anxiety about student well-being, surveillance offers school officials a chance to intervene when warning signs appear, as well as an accountability defense when threats go undetected. 2 See Fedders, supra note 1, at 1675, 1687-88. But these benefits often come with costs to students—especially those already disadvantaged. Classroom drafts are sometimes flagged before a student can fully contemplate their own words, 3 See Sharon Lurye, Students Have Been Called to the Office—and Even Arrested—for AI Surveillance False Alarms, Associated Press (Aug. 8, 2025), https://perma.cc/YVA7-6Z8T. and even a simple web search might generate an alert that sets an institutional response into motion. 4 See Ellen Barry, Spying on Student Devices, Schools Aim to Intercept Self-Harm Before It Happens, N.Y. Times (Dec. 9, 2024), https://perma.cc/YS4B-5B3Y. In extreme cases, students have been arrested over jokes 5 See Lurye, supra note 3. and sensitive student files have been exposed publicly when privacy safeguards proved insecure. 6 See Kendrick, supra note 1. Insofar as these harms disproportionately affect certain student populations, surveillance creates a tension between safety and equal access that affects the conditions under which public education is delivered.

The compulsory nature of public education complicates the legal ramifications of this tension. Many students lack the means to opt out of public school, and classroom participation often requires the use of school-managed devices and networks. 7 See Sharma, supra note 1; 105 Ill. Comp. Stat. 5/26-1 (2025) (requiring parents to send their children to public school barring certain exemptions). Surveillance is difficult to avoid for these students, a reality that can alter how they experience school. 8 See Danielle Keats Citron, The Surveilled Student, 76 Stan. L. Rev. 1439, 1448-52 (2024). A student under constant surveillance occupies a different educational environment from a student who is not. 9 See Sharma, supra note 1 (“The ability to learn and explore without a constant fear of monitoring, criminalization, and retaliation is critical to students in the public education system.”).

The available evidence suggests that these costs fall unevenly on students. For example, LGBTQ+ students face distinctive risks of disclosure of personal information to parents and referral to police. 10 See Elizabeth Laird, Hugh Grant-Chapman, Cody Venzke & Hannah Quay-de la Vallee, Ctr. for Democracy & Tech., Hidden Harms: The Misleading Promise of Monitoring Students Online 21 (2022), https://perma.cc/J53U-EQSY [hereinafter Hidden Harms]. And students of color and low-income students are more likely to depend on school devices and networks for things like routine internet access. 11 See id. at 23-24; see also Sharma, supra note 1; DeVan Hankerson, Cody Venzke, Elizabeth Laird, Hugh Grant-Chapman, Dhanaraj Thakur, Ctr. for Democracy & Tech., Online and Observed: Student Privacy Implications of School-Issued Devices and Student Activity Monitoring Software 10-11 (2021), https://perma.cc/M3S7-BP5P.

In a similar vein, disabled students are monitored more often and suppress their online expression at higher rates than their peers in response to surveillance software. That unequal burden presents distinct problems under Section 504 of the Rehabilitation Act, which forbids discrimination against disabled people in the provision of public services like education. 12 See Hidden Harms, supra note 10, at 23; Elizabeth Laird, Madeleine Dwyer & Hugh Grant-Chapman, Ctr. for Democracy & Tech., Off Task: EdTech Threats to Student Privacy and Equity in the Age of AI 22 (2023), https://perma.cc/2W5M-2SPV [hereinafter Off Task] (noting that disabled students are more likely than their peers to report being monitored at school). The Supreme Court has assumed without deciding that Section 504’s protections extend beyond intentional discrimination to at least some cases of disparate impact, and the federal courts of appeals have split over whether such claims are cognizable. 13 See Alexander v. Choate, 469 U.S. 287, 299 (1985); compare Payan v. L.A. Cmty. Coll. Dist., 11 F.4th 729, 737-38 (9th Cir. 2021) (allowing disparate-impact claims under the Rehabilitation Act and its implementing regulations), with Doe v. BlueCross BlueShield of Tenn., Inc., 926 F.3d 235, 241-43 (6th Cir. 2019) (holding that the Rehabilitation Act does not prohibit disparate-impact discrimination).

This Essay argues that automated student monitoring implicates Section 504 in two ways. The first relates to a school’s response after surveillance software flags a student’s expression. 14 Professor Danielle Keats Citron has examined student surveillance through privacy law and proposed regulatory reforms, while also noting that existing civil-rights law may support nondiscrimination duties. See Citron, supra note 8, at 1465-72. This Essay develops the specific implications of Section 504 for school districts’ use of automated monitoring. When an alert signals that a student may need special education or related services, officials who respond only with disciplinary measures or police referrals risk disregarding the evaluation and support duties that Section 504 imposes. A second problem stems from the systemic effects of pervasive monitoring on the educational environment. Section 504 guarantees disabled students meaningful access to a school’s educational program, and its implementing regulation extends to the methods through which a district administers that access. 15 See Rehabilitation Act of 1973 § 504, 29 U.S.C. § 794; 34 C.F.R. § 104.4(b)(4). By making monitored infrastructure the everyday medium of school participation, a district chooses such a method and therefore must account for whether that choice deters disabled students from using the channels through which they participate in school.

To address these civil-rights concerns, districts should deliberately consider both how alerts will be handled and how they are generated in the first place. School officials must ensure that responses to flagged speech comport with Section 504’s evaluation and support obligations. They should also assess whether the scope of the district’s monitoring program burdens disabled students’ use of school devices and networks and, if so, whether narrower approaches might adequately serve the same safety goals. Particular focus should be given to the kind of educational environment continuous monitoring creates for the students least able to avoid it.

This Essay proceeds in four parts. Part I traces the extensive use of surveillance software in modern public education. Part II examines the risks these systems pose to students, particularly to disadvantaged students. Part III contends that school districts may violate Section 504 if their responses to, or implementation of, surveillance software fails to consider their obligations to disabled students. Part IV offers recommendations for ensuring compliance with Section 504 in the adoption and use of surveillance software.

I. Automated Monitoring in Public Schools

Modern surveillance software promises early detection of threats to student safety. 16 See Barry, supra note 4. Hoping to ensure a secure learning environment and to direct resources toward at-risk students, districts closely monitor school-issued devices and school-managed accounts and networks. 17 See Sharma, supra note 1; Kendrick, supra note 1. Schools thus have constant eyes on the primary channels through which students complete their daily coursework and connect to school personnel. The software’s artificial intelligence (AI) systems scan student activity in these channels, including searches and messages, and use algorithms to flag words and phrases associated with self-harm or violence for further review. Students typically lack any choice over whether they are monitored. 18 See Bryan & Lurye, supra note 1 (describing Gaggle’s machine-learning scan of student searches and writing, and reporting that some families are unable to opt out of monitoring).

District officials have considerable control over how these systems connect student speech to institutional action. 19 See id.; Hidden Harms, supra note 10, at 11-12. Different districts running the same system may diverge in how they configure their alert settings. One district may route signs of distress to support staff, while another may send them directly to school police. Other choices include which staff receive alerts, in what order, and how long captured material is stored.

Districts adopt these AI surveillance systems without knowing what exactly they will deliver. Vendors’ data on how often alerts are triggered and what triggers them is usually not shared with the public or independent researchers, and districts often do not generate comparable data themselves. 20 See Barry, supra note 4 (reporting that vendors maintain data on their systems’ accuracy and that one district retained no data on alert outcomes). Whether surveillance software reliably identifies students in danger therefore remains an open question. Despite uncertainty over the software’s efficacy, schools are making substantial financial investments in surveillance. For example, Vancouver, Washington’s three-year contract with Gaggle cost approximately as much as employing one additional counselor. 21 Bryan & Lurye, supra note 1.

These systems also complicate questions of accountability regarding who exercises judgment over student expression. Software vendors design the systems, giving private corporations a substantial role in deciding what student speech warrants official attention and the parameters within which that speech is captured and stored. 22 See Sharma, supra note 1. Even so, school officials remain legally responsible for the conditions under which students participate in school, regardless of whether those conditions are substantially shaped by the companies with whom they contract.

II. From Writing to Warning

As surveillance systems gain prevalence in public education, accounts of their failures are mounting. In Fairview, Tennessee, a 13-year-old eighth grader made an offensive joke in a monitored school chat that the district’s system flagged. School administrators notified the police, who strip-searched the student and jailed her overnight. 23 Lurye, supra note 3. In Lawrence, Kansas, the district’s system generated more than 1,200 alerts in ten months; nearly two-thirds were ultimately deemed nonissues, including more than 200 triggered by homework assignments. 24 Id. Students in one photography class were summoned over images the system identified as nudity and automatically deleted from their Google Drives, but backup copies showed that the alerts were false. 25 Id. The district’s software also allegedly monitored the work of student journalists, including publication drafts and collaborative emails. 26 See Grace Hills, Lawrence School Board’s Spyware Renewal Leaves Student Journalists’ Concerns Unresolved, Kan. Reflector (July 16, 2024, 11:12 AM CDT), https://perma.cc/7JXA-5S7X (describing student journalists’ allegations that monitoring interfered with their reporting); Natasha Torkzaban, Morgan Salisbury, Maya Smith & Jack Tell, Fighting for Our Rights: Gaggle & USD 497, Budget (Apr. 18, 2024), https://perma.cc/8LW9-TBVZ (describing students’ challenge to the district’s monitoring of journalism materials). And in Vancouver, Washington, student writing about personal issues like sexuality and heartbreak was flagged and stored by the district’s monitoring software. 27 See Bryan & Lurye, supra note 1. A public-records response later released nearly 3,500 sensitive student documents without redaction or password protection. 28 Id.; Kendrick, supra note 1.

These outcomes stem from how surveillance systems are currently designed and deployed. By filtering for language about emotional distress or self-harm, the software draws crude distinctions on the basis of content. 29 See Bryan & Lurye, supra note 1 (describing Gaggle’s algorithmic detection of potential indicators of self-harm and suicide). Student speech that would once have passed unnoticed thereby becomes institutionally legible, and therefore institutionally actionable. That the systems can obscure crucial context only amplifies the concern. Teachers report that surveillance systems are used more often for disciplinary enforcement than for mental-health intervention. 30 See Hidden Harms, supra note 10, at 12 (reporting that teachers identified disciplinary violations as a leading purpose of monitoring more often than mental-health crises and reported more monitoring-related disciplinary consequences than counseling referrals).

For students, this kind of system can chill a wide range of inquiries. 31 See Neil M. Richards, The Dangers of Surveillance, 126 Harv. L. Rev. 1934, 1950-52 (2013) (noting that surveillance can chill intellectual activities). Students who expect quick searches or unfinished writing to become official evidence may engage with their assignments more cautiously. That pressure is especially acute when students are still developing their intellectual faculties. A private chat can be logged in a student’s file before they have time to rethink what they have typed.

These costs affect different groups of students to varying degrees. Students with physical disabilities and learning differences report greater expressive restraint under monitoring, 32 See Hidden Harms, supra note 10, at 23; Nicole Fuller & Lindsay Kubatzky, Addressing the Disproportionate Impacts of Student Online Activity Monitoring Software on Students with Disabilities, Fed’n Am. Scientists (June 25, 2024), https://perma.cc/TD65-ASMM. and special-education teachers report higher rates of law-enforcement contact following alerts. 33 See Off Task, supra note 12, at 29. 31 percent of LGBTQ+ students reported that they or another student had been contacted by a police officer or other adult over concerns about a crime following monitoring, compared with 19 percent of non-LGBTQ+ students. 34 See Hidden Harms, supra note 10, at 21. Moreover, roughly 60 percent of Black students, 60 percent of Hispanic students, and 70 percent of low-income students rely on school-provided devices. 35 See id. at 23.

Research on schools that use cameras and metal detectors finds that Black students are four times more likely to attend the most heavily surveilled schools. 36 See Odis Johnson Jr. & Jason Jabbari, Infrastructure of Social Control: A Multi-Level Counterfactual Analysis of Surveillance and Black Education, 83 J. Crim. Just., Nov.-Dec. 2022, at 1, 2; Hidden Harms, supra note 10, at 24. The surveillance index in the Johnson and Jabbari study covers cameras, metal detectors, dress codes, and similar measures. Separate survey data suggests that those disparities extend to monitoring-related discipline. 55 percent of Hispanic students reported that they or someone they knew had gotten into trouble because of monitoring, compared with 41 percent of white students. 37 See Hidden Harms, supra note 10, at 24. Collectively, this data indicates that the students with the fewest practical ways to avoid monitored infrastructure are often the students most likely to bear its costs.

III. Section 504 and Monitored Participation

For disabled students, automated monitoring implicates Section 504 at two stages. 38 The other disparities described in Part II may also raise questions under Title IX and Title VI, whose requirements differ from Section 504’s. See 20 U.S.C. § 1681(a); 42 U.S.C. § 2000d. A district may mishandle expression after monitoring brings it to official attention, particularly when an alert conveys a need for student evaluation or support. Legal concerns can also arise when the prospect of monitoring changes how disabled students use the accounts through which they participate in school.

A. Responding to Alerts

Under Section 504 and its implementing regulation, schools receiving federal funds must provide a free appropriate public education (FAPE) to qualified students with disabilities. 39 34 C.F.R. § 104.33(a)-(b)(1) (requiring recipients operating public elementary or secondary education programs to provide a free appropriate public education consisting of services designed to meet disabled students’ individual educational needs as adequately as the needs of nondisabled students). Schools must also conduct an evaluation of any student who, because of disability, needs or is believed to need special education or related services, both before an initial placement and before any subsequent significant change in placement. 40 Id. § 104.35(a).

How an automated alert bears on those duties depends on what it reveals and what the school already knows. Whereas a single search for self-harm resources may be ambiguous, a pattern of searches that conveys distress could give the district reason to believe the student might need such services. For a student already protected by Section 504, the same material may bear on whether a proposed removal or other disciplinary response requires evaluation and coordination.

In 2022, the Department of Education’s Office for Civil Rights (OCR) identified surveillance technologies, student searches, law-enforcement referrals, and threat or risk assessments as among the discipline-related activities subject to Section 504. 41 U.S. Dep’t of Educ., Off. for C.R., Supporting Students with Disabilities and Avoiding the Discriminatory Use of Student Discipline Under Section 504 of the Rehabilitation Act of 1973, at 3 (2022), https://perma.cc/X2H8-5L9V [hereinafter Supporting Students] (identifying surveillance technologies, student searches, law-enforcement referrals, and threat or risk assessments as among discipline-related activities subject to Section 504). The guidance states that it is nonbinding and does not create or impose new legal requirements. Id. at i n.3. The guidance explained that behavior suggesting a student may need special education or related services can trigger an evaluation duty. Where a threat-assessment team fails to coordinate with a student’s Section 504 team, OCR warned, the result may be a denial of FAPE. 42 See id. at 14-15 (explaining when discipline constitutes a significant change in placement); id. at 21-22 (warning that failure to coordinate threat assessment with a student’s Section 504 team may deny a free appropriate public education).

Some flagged expression might require both a threat assessment and a supportive inquiry. Writing about self-harm, for example, may prompt disciplinary review even though it also gives the district reason to consider whether the student needs services. What Section 504 requires will depend on the circumstances. But when an alert gives officials reason to believe that a student may need services, responding with discipline alone disregards that evidence. 43 See id. at 6 (explaining that a district may not ignore evidence that a student may need special education or related services nor unreasonably delay evaluation).

B. Mandatory Monitoring and Meaningful Access

The second Section 504 problem arises from widespread monitoring more generally. When districts require students to use surveilled school networks and devices, the terms on which those students participate in school change. In other words, the burden might result from the mere fact of constant surveillance.

Indeed, survey evidence suggests that disabled students experience pervasive monitoring differently than do other students. For example, in 2022, 67 percent of respondents with physical disabilities and 60 percent with learning differences reported withholding their true thoughts or ideas due to monitoring, as compared with 46 percent and 45 percent of their respective peers. 44 See Hidden Harms, supra note 10, at 22-23. A later survey found that students with individualized education programs or Section 504 plans reported being subjected to school monitoring at a higher rate than other students, 89 percent to 78 percent. 45 Off Task, supra note 12, at 22. The 2022 survey covered students in grades nine through twelve at schools using activity-monitoring software, and the relevant subgroup samples were forty-eight and ninety-two students. Together, these findings suggest unequal exposure to monitoring and disproportionate attendant self-censorship among disabled students.

To put the disparity in more concrete terms, consider a hypothetical student who suffers from a mixed anxiety and depressive disorder that substantially limits her ability to concentrate in class. The student must use her school-issued account to access the internet and to message school officials. But because she knows that language associated with self-harm will trigger her school’s surveillance software and could expose her private thoughts to administrative personnel, she avoids using the account to contact a counselor about her suicidal ideation or to search for guidance on how to address her intrusive thoughts so that she can better focus in class. Her ability to use her account is thus constrained by the school’s surveillance system.

Whether that constraint constitutes a denial of access turns on what benefit the district has made available to all students, and on what stands between her and that benefit. The Supreme Court considered a similar question four decades ago in Alexander v. Choate. 46 See Choate, 469 U.S. at 289. Tennessee had cut annual inpatient coverage under its Medicaid program from twenty days to fourteen days due to budgetary shortfalls. 47 See id. Disabled Medicaid recipients brought suit in response, arguing that the cut disproportionately affected them because they were more likely to need extended hospital stays. The Court assumed without deciding that Section 504 prohibits at least some disparate effects, pointing to the law’s core concern with remedying widespread careless neglect of disabled people’s needs. 48 See id. at 289-90, 299. But the Court rejected the challenge at hand, on the grounds that the disabled recipients remained eligible for the same effective hospital services as all other recipients. 49 See id. at 299, 302. The Court held that Tennessee could define the benefit it offered as long as the state’s definition did not effectively deny disabled recipients meaningful access. 50 See id. at 301, 303.

The relevant benefit for surveilled students is the coursework and the support services that a district delivers through its monitored infrastructure. In Choate, Tennessee’s fourteen-day cap applied without regard for why a patient was hospitalized. 51 See id. at 302 n.22. The Court concluded that the same effective hospital services were available to all recipients within that limit, as the cap restricted only how much of the benefit a recipient could receive. Lower courts have subsequently distinguished limits on the scope of a benefit from policies that obstruct access to a benefit already provided. In American Council of the Blind v. Paulson, for example, the D.C. Circuit held that the uniform size and texture of paper currency presented such an obstacle, denying blind users meaningful access to the existing currency system. 52 See Am. Council of the Blind v. Paulson, 525 F.3d 1256, 1266-67 (D.C. Cir. 2008).

Under this reasoning, a district’s surveillance system can operate as such an obstacle. To seek counseling about her intrusive thoughts, the hypothetical student must communicate the very subject matter that the system is configured to flag. She therefore risks exposing her private thoughts to school personnel (other than the counselor) if she uses the school’s monitored account. Where that prospect materially deters her from getting support, the monitored channel may be less effective as a means of obtaining the counseling service the district already provides.

Section 504’s implementing regulations similarly prohibit recipients from affording disabled students an unequal opportunity to obtain a benefit or a service less effective than what other students receive. 53 34 C.F.R. § 104.4(b)(1)(ii) (prohibiting a recipient from affording “an opportunity to participate in or benefit from the aid, benefit, or service that is not equal to that afforded others”); id. § 104.4(b)(1)(iii) (prohibiting a recipient from providing a service “that is not as effective as that provided to others”); id. § 104.4(b)(2) (providing that equally effective services need not produce identical results but must afford an equal opportunity to obtain the same benefit). That includes equal opportunity in nonacademic services like counseling. 54 See id. § 104.37(a)-(b) (requiring equal opportunity in nonacademic services and identifying counseling services, health services, and referrals as among the covered services). When surveillance discourages a student from using a school account to seek counseling or other support, it risks denying the essential equality of opportunity that the regulations require. And a separate provision bars a recipient, directly or through contractual arrangements, from using methods of administration that have the effect of subjecting disabled students to discrimination. 55 See id. § 104.4(b)(4) (prohibiting a recipient, “directly or through contractual or other arrangements,” from using criteria or methods of administration that have discriminatory effects or substantially impair program objectives “with respect to handicapped persons”). A surveillance system designed by a vendor and configured by the district is such a method.

These obligations bind districts as recipients of federal funds, although courts disagree about whether effects-based regulations such as § 104.4(b)(4) are independently enforceable in private suits. 56 Compare Ability Ctr. v. City of Sandusky, 385 F.3d 901, 913-14 (6th Cir. 2004) (allowing private enforcement of an Americans with Disabilities Act (ADA) Title II regulation that effectuated the statute’s meaningful-access mandate), with Three Rivers Ctr. for Indep. Living, Inc. v. Hous. Auth., 382 F.3d 412, 425-31 (3d Cir. 2004) (declining to permit private enforcement of systemic regulatory obligations extending beyond rights created by Section 504 itself). Whatever the available enforcement mechanism, surveillance software that denies disabled students an equal opportunity to obtain educational benefits raises legal concerns under Section 504 and its implementing regulations.

IV. Front-End Accountability Before and During Deployment

District officials should regularly evaluate their use of surveillance software. Ideally, officials will thoroughly consider the benefits and costs to students before implementing surveillance systems. A district that adopts monitoring after a serious incident may face strong political headwinds if it later tries to cancel its contract. The obligations described above therefore make acquisition approval the most opportune moment to carefully consider whether pervasive monitoring is in fact needed to make schools safer and, if so, what scope of monitoring is needed to achieve the school’s aims. And even in the face of public pressure, contract renewals also offer districts an opportunity to assess their systems’ efficacy and drawbacks.

In response to concerns over inequities that surveillance software might create, districts might reasonably answer that a school that misses a credible threat of violence risks litigation and public outrage, while its students face worse. That pressure is surely substantial, and civil-rights law leaves districts ample leeway when it comes to preemptive safety measures. Districts may also contend that monitoring advances Section 504’s aims rather than straining them, as an alert can identify a troubled student who was not previously on officials’ radar.

A district that defends its system as a way of finding students who need help, however, has conceded that its alerts offer insight into its students’ mental health. Treating those same alerts simply as disciplinary material would thus disregard the district’s statutory response obligations. The same defense also implicates the district’s more general obligation to ensure that the benefits it offers are equally accessible. A system justified by its capacity to identify students who need help must be evaluated on whether it drives those students away from the channels through which they might obtain help.

Determining whether a monitoring system creates the effects described above requires evidence of how the system actually performs. A district should analyze how often alerts identify actual danger or sweep harmless expression into review. It should likewise assess whether disabled students are more likely to face discipline after an alert or to avoid monitored channels before one is generated. That information should shape a formal plan for responding to alerts. For example, districts should decide in advance who receives alerts that might suggest a need for special education or related services. They should also specify what must occur before an alert is used for disciplinary purposes. To be sure, emergencies require flexibility to immediately respond to pressing danger. But absent an immediate threat, districts should ordinarily involve support personnel before turning to disciplinary measures. In terms of potential accommodations, a student with a documented disability might receive an unmonitored channel for reaching counseling staff, or a firewall that keeps searches linked to the student’s disability out of the student’s disciplinary record.

Even a proactive response plan can only go so far, and districts should also carefully calibrate the scope of their monitoring. Continuous review of mundane student activity is harder to justify than monitoring tied to a particularized concern. For that reason, districts should look closely at how much student activity their systems review and how long flagged material is retained. And they should determine whether narrower monitoring or greater investment in school personnel would address the same safety concern, while simultaneously reducing the pressure on disabled students to avoid school networks or devices. 57 See Barry, supra note 4 (noting that Baltimore scaled back its monitoring program to school hours only, in response to a large number of false positives).

Conclusion

Districts are responsible for the educational environment that results from the surveillance software the districts implement. In adopting comprehensive monitoring systems, districts thus make choices that carry civil-rights consequences. They should therefore be prepared to explain why a system operating at a particular scale is warranted and how captured student expression will be handled. They must also confront whether disabled students can participate in the resulting educational environment on equal terms.

*Assistant Professor of Law, Loyola University Chicago School of Law. I am grateful for the feedback I received from the participants of the Loyola University Chicago summer faculty workshop.