Skip to content

Cornell, criminal justice drift, and the case for taking a position on sanctions

4 October 2026 — Commentary

Cornell University campus at sunset.
Photo by Christopher Chan on Unsplash

Content note: this piece discusses a reported sexual assault and the university and court processes that followed. It does not describe the assault. Support links are listed at the end.

Contents

Cornell University is facing scrutiny over how it responded to a student’s report of sexual assault. In November 2024, an undergraduate reported that she had been sexually assaulted by members of a fraternity the month before. Cornell investigated seven students under its Title IX policy. In September 2026, the reporting student brought a civil claimagainst the university, the seven students, and others. Since then, the criminal investigation has been reopened, the state Attorney General has been appointed as special prosecutor, and Cornell has agreed to an independent review of how it handled the case. According to press reports, the responding students have denied wrongdoing. On 3 October 2026, Cornell’s president said the university “must do better” and would “be more transparent”.

Separately, one of the seven students asked a New York court to overturn the university’s finding against him. The court dismissed his petition on 23 September 2026, and he has appealed. Redacted documents from the university process became public through that case.

What we can and cannot know

I have read many of the publicly available court documents, Cornell’s published policy and statements, and press reporting, but I acknowledge that this represents a small part of a large record. Most of the court file is sealed and Cornell says federal privacy law prevents it from giving individual outcomes. Accounts of what was shared with the police differ, and criminal and civil proceedings are ongoing.

I do not seek to comment on what happened that night or on whether any outcome was right. The student who reported has left the university and is now seeing her experience discussed in public. The students who were investigated are also being discussed publicly. What the public record does show is a process that raises questions that many of us involved in student casework may recognise.

The pressure on universities

The scale of the Cornell process gives a sense of what universities are now frequently managing. The process involved more than 50 interviews, a 12-day hearing with seven responding students, a 1,732-page transcript, a 99-page decision, an internal appeal, and then a court challenge.

Universities are rightly expected to support the student who reported, treat responding students fairly, meet regulatory requirements, withstand legal challenge, and answer to their communities and the press. This is difficult work and the staff and panel members involved are likely to have been doing their best within the rules they worked to. But this case also shows how much harm can follow when a process is not done carefully and well.

Where the process may have added to the harm

Several features of the process, as they appear in the public record, may have added to the distress of the people involved.

  • Time. About 14 months passed between the first report and the final university decision. The appeal decision, which is about two pages long, came nearly five months after the panel’s decision. (This is likely to reflect the volume of material and the complexity of the process, and may have been unavoidable.)
  • Questioning. Under the US federal rules that applied, each party’s adviser could cross-examine the other parties. One responding student’s lawyer questioned the student who reported over two days, in a hearing with seven responding students.
  • Attention. During a witness’s evidence, one panel member took part for 45 minutes while driving. The chair dealt with it at the time, and the court saw no basis to fault her decision. It may still have affected the parties’ confidence in the process.
  • Notice. The responding student who later went to court was found responsible for attempted sexual exploitation, which had not been specifically put to him as a named allegation. The court accepted that he had notice of the alleged facts, but his complaint was about voice: knowing the case you have to answer. Cornell’s procedures now say that an attempt can be found without being specifically alleged.
  • Trust. Cornell’s online FAQ said it had shared all investigatory information with the District Attorney, but this wording was later changed. The president said he had heard frustration that Cornell “has not communicated enough”, which “has led to distrust”.

Research on institutional betrayal suggests that when an institution fails to prevent harm, or to respond well, this can add to the trauma of the original experience (Smith and Freyd, 2013 and 2014). In one US study, every survivor interviewed saw some part of their reporting experience as negative, and for many who reported the Title IX investigation caused further harm (Lorenz and others, 2024). Responding students also report negative emotions during these processes (Roskin-Frazee, 2022). In another small study, survivors said that seeing others treated badly in the news would make them less likely to report (PettyJohn and others, 2023).

Criminal justice drift in this case

Cowan and Munro (2021) used the phrase “criminal justice drift” to describe UK university responses to sexual misconduct becoming more like criminal justice processes. They argue that this may make it harder to develop processes that are tailored to students’ needs and trauma-informed. The Cornell record contains several examples.

  • Structure. A 12-day hearing, lawyers cross-examining, a lawyer as chair, an internal appeal, and then a court challenge. The court decision refers to “charges” throughout, and many university codes use this term.
  • Definitions. New York’s notice rules for these cases rely on federal criminal definitions that require physical contact. The court found that those rules therefore did not cover sexual exploitation by observing someone without their consent.
  • Reasoning. The court compared the finding with criminal law, where an attempt is a “lesser included offense” of the full crime. It reasoned that if a criminal defendant, “whose liberty interests are at stake”, can be convicted of an attempt without being charged with it, the university’s finding could not be “arbitrary and capricious”.

The comparison had two effects: it treated the university process as similar to a criminal one, and it used the difference between them to limit what the university had to do. The same tension appears in public discussion. Some commentary has compared Cornell’s sanctions with what a criminal court might have done, while Cornell’s FAQ says its process “applies university policies and standards, not criminal law”.

UK processes differ in several ways. Universities usually pause their own process during a police investigation, questions are often put through the chair, and the standard of proof is the balance of probabilities (OIA Good Practice Framework).

Criminal justice drift can also affect the position of the student who reported. Bull, Calvert-Lee and Page (2021) suggest that UK disciplinary processes used for student complaints about staff sexual misconduct are “too often modelled on the criminal justice system”. In their view, this leaves the student who complained in the position of a witness, with no right to see the evidence or to know the full outcome. They say the same problem exists in cases between students, although it tends to be greater in staff cases. They propose a process “more akin to civil justice”, with equal rights for both parties. At Cornell, the federal rules made the student who reported a party to the process, with her own adviser.

The responding student who went to court did not argue that the evidence failed to support the panel’s decision. His challenge was based on procedure: he argued only that he had been found responsible for something he had not been charged with. He lost, and he has appealed.

For universities, challenges like this take time, cost money, and bring scrutiny. Cornell defended its decision by pointing to the wording of its formal complaint and its written procedures.

For the student who reported, the effects may be more difficult. The university process ended in January 2026, but her experience is still being examined in court and in the press. Her evidence, and messages she sent at the time, are now discussed publicly, and she has no control over that. A legal challenge to a civil process can mean going through parts of it again.

In the UK the routes for challenge are different: a complaint to the OIA (England and Wales), the SPSO (Scotland), or NIPSO (Northern Ireland); a contract claim; or, for some providers, judicial review. Being held to a legal standard during a civil process can feel adversarial and intimidating for everyone involved. It is also a reason for institutions to be clear, before any case arises, about what they are trying to achieve.

How sanctions are decided

Cornell’s procedures list the factors a panel will consider when deciding a sanction. These include the severity and circumstances of the conduct, its impact on the people involved and on the community, any previous misconduct, and “any other mitigating, aggravating, or compelling factors”. The sanctions available range from an oral warning to expulsion. The procedures do not say which outcome a student, or the wider community, should expect for which kind of finding.

In the decision that is now public, the panel imposed a no-contact order and educational sanctions, including a reflection paper; these are also common outcomes in UK higher education. It noted that the responding student who appealed had no previous non-academic disciplinary record and said it intended the sanction to give him an “opportunity for understanding and maturation”. That reflects a choice about what sanctions are for. I am not suggesting here that the choice was wrong.

There have been claims that students only had to write essays, which Cornell denies. Cornell’s Student Assembly has asked the university to disclose the sanctions given and the status of the students’ degrees. CBS News reported that the president told student representatives that one of the seven had already graduated before any punishment could be given.

That raises a question that comes up in the UK too. If a responding student who is close to graduating is found to have committed serious sexual misconduct, what should happen? The OIA notes that “if a responding student is about to graduate the provider will not be able to begin a disciplinary process after that point”. In one OIA case, a student was awarded a degree in error while a matter was unresolved, and the OIA recommended that the university’s later decision to expel them be overturned.

Deciding case by case can feel fair, because cases differ. But research suggests that sanctioning decisions can be affected by factors that institutions may not intend to weigh. In one vignette study, university staff and students rated expulsion as less appropriate when the perpetrator was intoxicated than when they were sober (Wesenberg and Jung, 2026). In an ethnographic study of US Title IX administrators, Bedera (2024) found that they defined their primary role as “protecting the futures of young men”. Without a stated position, decisions made case by case, and with reference to what was done before, may repeat or uphold older assumptions and inequalities.

Consistency is important, and the OIA already expects it. Its guidance says providers should apply penalties consistently, for example by keeping anonymised summary records of offences, mitigating factors, and penalties. It also says procedures should indicate the penalties likely to apply in different circumstances.

However, I believe consistency should mean consistency with a position the institution has carefully considered, and stated, not simply with what it has done before.

Setting out a principled position

Each institution needs to decide what it is trying to achieve when it sanctions a student for sexual misconduct. Is the priority the safety of the reporting student and the community? Accountability? Education and change? How much weight, if any, should the institution’s reputation carry? What does a good outcome look like?

A stated position might cover:

  • the purposes of sanctions, and how they are balanced
  • the outcomes the institution would usually expect for different kinds of finding
  • how factors such as intoxication, previous record, and stage of study are treated
  • what happens when a responding student is close to completing their course
  • what the reporting student, and the wider community, will be told
  • what anonymised data will be published.

This position should not be private. It should be published and supported openly by senior leaders. Other bodies already do something similar; the Medical Practitioners Tribunal Service, for example, publishes guidance for its tribunals that explains what sanctions are for and indicates expected ranges of outcome, while leaving room for judgement. I note that its tribunals have a different role from a university disciplinary panel; they decide whether a doctor is fit to practise, and the guidance says sanctions are meant to protect the public, not to punish, although they may have a punitive effect. However, I would suggest that this kind of transparency about the purpose of sanctions and outcomes is good practice.

Transparency about outcomes is possible within privacy law. New York already requires colleges to report to the state each year how many students were found responsible, with “a description of the final sanctions imposed”, and the aggregate data is published. In the UK, the OIA says data protection law does not prevent providers from sharing outcomes with reporting students, although they must balance this with the responding student’s privacy. In interviews with people who reported to UK universities between 2016 and 2021 (before the OIA guidance was published), Bull (2026) found that in formal processes the “outcome” was a finding for or against the responding party, “alongside sanctions of which the reporting party is usually not informed”. She argues that, although there is broad agreement that some sanctions can be shared with reporting parties, this is not yet happening in practice. Smith and Freyd (2014)suggest that one of the first steps an institution can take to repair betrayal is a careful self-study. That could include looking openly at its own past decisions about sanctions and about sharing outcomes.

A framework won’t remove the need for judgement, and it needs to be fair and trauma-informed for responding students as well as reporting students. But an institution that has carefully thought through and agreed its position, and explained it openly to its community, is likely to find its processes easier to apply in practice, its decisions easier to defend when they are questioned, and easier to explain to the people affected by them.

Questions to ask in your own institution

  • Have we decided, and written down, what our sanctions are for?
  • Would a panel member know what outcome we would usually expect for a given finding, and why?
  • What happens if a responding student is close to completing their course?
  • What will we tell the reporting student, and the wider community, about outcomes?
  • When did we last review our past outcomes for patterns?
  • Has our leadership said publicly that it supports this position?

If you need support

In the US

Where you study

Your university or college may also be able to help, through its student support or wellbeing service, your students’ union, or, in the US, its Title IX office.

← Back to blog

Ready when you are

Dealing with something similar?

Arrange a no-obligation call, or email us directly. We reply within two working days.

Get in touch