What Exactly Are They Arguing Against?
A critical reading of Adam Thompson, “Should Universities Have a Statutory Duty of Care?”, Fireside Speech, 13 January 2026.
Author: Robert Abrahart
Framing the question
On 13 January 2026, MPs met in Westminster Hall to debate the potential merits of a statutory duty of care for universities. Adam Thompson, Labour MP for Erewash, was not called to speak due to time constraints and subsequently delivered his intended contribution in an online video titled Should Universities Have a Statutory Duty of Care? [YouTube] [Substack]
The recording provides a useful opportunity for critical reading. Thompson sets out a concise case against introducing a statutory duty, drawing explicitly on arguments advanced by two key representative bodies within higher education: Universities UK (UUK), representing institutions, and the University and College Union (UCU), representing staff. At the outset, he properly declares his own relevant interests: he chairs the All-Party Parliamentary Universities Group, is a member and former workplace representative of UCU, and retains an unpaid honorary associate professorship at the University of Nottingham.
Before considering the substance of his objections, however, there is a more fundamental difficulty. The Westminster Hall debate concerned the potential merits of a statutory duty of care. That wording did not prescribe one particular model. It invited consideration of what such a duty might look like, what it might achieve, and what advantages or disadvantages different formulations might bring.
Thompson does not undertake that exercise. Instead, his contribution is concerned almost entirely with perceived disadvantages. Such concerns may deserve consideration, but there is no corresponding exploration of whether some form of statutory duty might have any potential merits, what those merits might be, or how they should be weighed against the objections he raises.
More importantly, the statutory duty itself remains undefined. Thompson describes a new “overarching statutory duty of care”, but does not explain what form it might take, when it might arise, what responsibilities it might encompass or what standard of conduct might apply. Without that clarity, it is difficult to know precisely what his objections are directed against.
An Argument Inherited from 2023
Prior to the first Westminster Hall debate on student duty of care in 2023, Universities UK published two parliamentary briefings outlining its opposition [April 2023 briefing] [May 2023 briefing].
Three years later, the case advanced by Thompson remains remarkably close to the sector's original position on a statutory duty: universities already owe students a limited general duty of care “not to cause harm by careless acts or omissions”; other legal and regulatory frameworks already apply, including data-protection requirements; students are autonomous adults; universities exercise limited control over their wider lives and are not in loco parentis; there is a risk of regulatory overreach; and additional legal obligations may encourage defensive, compliance-led practice.
There are differences of emphasis. Most notably, Thompson also draws on UCU concerns about resources, staff workloads, pressure on mental-health support services, insecure employment, confidentiality and the danger of responsibility for complex risks being displaced onto individual members of staff. Repetition does not make an argument wrong. But three years later, and in a debate expressly concerned with potential merits, there was an opportunity to revisit the underlying assumptions rather than simply repeat the sector’s earlier arguments.
What Does “Cause Harm by Careless Acts or Omissions” Mean?
At the centre of Thompson’s argument is the limited general duty. That apparently simple formulation conceals an important distinction, best understood by separating three situations:
Situation 1: Causing Harm by a Careless Act
The first is straightforward. A university does something carelessly and, as a result, causes harm to a student. This might happen, for example, when carrying out an assessment, operating a procedure or making a decision. In legal terms, the question is whether the university exercised reasonable care and skill in what it did.
Situation 2: Causing Harm by a Careless Omission
The second situation involves an omission, but it is important to be precise about what that means. Suppose the university already has a responsibility to do something and fails to do it. If that failure causes harm, the omission may be relevant to negligence. But the sequence matters: the responsibility to act must come first. A failure to act does not become negligent merely because, with hindsight, it can be described as an “omission”. The word “omission” cannot itself establish the responsibility that makes the failure to act potentially negligent.
Situations 1 and 2 therefore share an important feature. In both, the university causes harm through the careless performance or non-performance of something for which it is already responsible. The underlying proposition is simply not to cause harm through careless conduct. In broad terms, this reflects a negative duty: do no harm. That is a substantially different proposition from imposing a responsibility to protect students from being harmed in other ways.
Situation 3: Harm Arising from Other Sources
The third situation raises a different question. Here the potential harm arises from another source and is not something the university itself caused through careless conduct (e.g. Rosen and Martinez). This is where a positive duty may become relevant: an obligation to take reasonable action.
Consider a simple hypothetical. A university knows that a student is suicidal. It did not cause the underlying suicidality. What responsibilities, if any, arise from that knowledge?
Two different questions may now need to be considered. First, what does the university itself decide to do, or continue doing? Conduct that might ordinarily present little or no risk could have very different consequences when applied to someone whom the institution knows to be particularly vulnerable. The university may not be the source of the potential harm, but its subsequent conduct could nevertheless increase the risk that harm will occur. In those circumstances, reasonable action may require it to alter or discontinue the activity concerned. Second, what does the university decide not to do, or fail to do? If it takes no action in response to a risk of harm arising from another source, its inaction cannot simply be labelled a careless omission. The prior question is whether, in the circumstances, it had a positive responsibility to take reasonable steps.
An example from another jurisdiction illustrates how the law might approach such a situation where a university becomes aware that a student has expressed an intention to take their own life or has attempted to do so. In Nguyen v Massachusetts Institute of Technology, the Massachusetts Supreme Judicial Court identified a limited duty that could arise where a university had actual knowledge of specified indicators of suicide risk. Where that duty was triggered, reasonable measures included initiating an existing suicide-prevention protocol; contacting appropriate university officials to assist the student in obtaining clinical care; if the student refused such care, notifying their emergency contact; and, in an emergency, contacting police, fire or emergency medical personnel. The court considered that such measures, where appropriate to the circumstances, would be reasonable and would satisfy the duty. The court emphasised that this limited duty respected the privacy and autonomy of adult students and recognised the capabilities of university staff who were not clinicians. The duty was not, however, found to have arisen on the facts of Nguyen's own case.
Nguyen therefore illustrates both the possibility and the limits of a positive duty. Knowledge does not automatically establish a positive legal duty or make the university responsible for preventing suicide. But what was known, by whom, the foreseeability of harm, the nature of the relationship, any responsibility already assumed and the reasonable steps available may all be relevant. The question is what, if anything, the university was reasonably required to do — or avoid doing — in the circumstances.
Does a Duty of Care Require Control?
Thompson's argument then moves from causing harm to a rather different proposition. Students are adults, universities are not in loco parentis, and a statutory duty, he suggests, could imply “a level of control universities don't and can't possess”. But why would a statutory duty necessarily require that level of control?
Thompson’s assumption that a statutory duty would require a high degree of control over students can be traced to the two sector briefings discussed earlier. In both, the limited general duty forms part of their argument against the imposition of an additional duty. That additional duty is characterised as one supposedly to “prevent self-harm” or “prevent suicide”, leading to comparisons with the kind of control associated with prisons or certain hospital settings.
That is a much more demanding conception of a statutory duty. Responsibility need not mean guaranteeing that harm will be prevented. As Nguyen illustrates, even a positive responsibility to take reasonable action may require only limited steps: initiating an established protocol, involving appropriate university officials, helping a student obtain appropriate clinical intervention, notifying an emergency contact where appropriate, or contacting emergency services in an emergency. None requires the university to take control of the student's life.
Nor does adult autonomy resolve the issue. It does not preclude ordinary responsibilities to warn of danger or avoid causing harm, and particular circumstances may require something more. Indeed, being over 18 changes both the nature of vulnerability and the protection required. Adult students cannot simply be protected through parental authority or ordinary child-safeguarding arrangements. They can nevertheless be exploited, harassed, sexually abused or otherwise harmed by other adults.
Condition E6 provides a useful illustration. The Office for Students imposes defined responsibilities on higher education providers concerning harassment and sexual misconduct, including requirements relating to prevention, reporting, investigation and decision-making. These responsibilities apply in an environment populated overwhelmingly by autonomous adults. They do not depend upon universities controlling those adults' wider lives.
Nor is control an all-or-nothing concept. Universities may have limited control over what students do, but considerable control over their own institutional activities and processes. Teaching is only one example. Universities conduct assessments, determine reasonable adjustments, operate disciplinary and complaints procedures, investigate allegations, make decisions affecting progression and continuation, provide accommodation and other services, and collect and act upon information about students.
Academic conduct provides a particularly useful illustration. As Dr Chelle Oldham argues in her recent HEPI article, “a conduct officer is not a clinician”. The officer's responsibility is not to diagnose or treat the student, but to operate the university's own process appropriately. As Oldham puts it, legal responsibilities “apply to the processes we run” as well as to teaching. That responsibility concerns the processes the university controls, not control over students' lives.
Similarly, a university may acquire information about a disability, sexual misconduct or a risk of harm without acquiring a high degree of control over the student concerned. That knowledge may nevertheless affect how its responsibilities should be carried out. Conduct that might be reasonable without that knowledge may require reconsideration once relevant information is known.
Does a Duty of Care Mean More Services?
Thompson’s use of the UCU argument raises another important question. His concerns about staff workloads, resources and responsibility being displaced onto individual members of staff are readily understandable. But they appear to assume that a statutory duty would require universities to provide more or better support services using an already stretched workforce. That concern deserves serious consideration, but it again depends upon what the proposed duty actually requires.
Thompson’s final spoken words are revealing. Having discussed a statutory duty of care, he asks whether universities should have such an additional legal obligation “to provide their support”. But that is a different question. A statutory duty of care in negligence need not require a university to provide any particular form of support or any specified services. If a university has undertaken to provide something, however, the question may instead be whether the law should require that institution to exercise reasonable care and skill in doing so.
This, then, is another example of the confusion surrounding the debate. One question concerns what universities should be required to provide. Another concerns the legal standard that should apply to the delivery of those services. The issue of whether that same standard should apply extends far beyond student services to assessment, reasonable adjustments, disciplinary proceedings, complaints, investigations, decision-making and other areas of university activity.
Clearer institutional responsibility might also help staff rather than add to their workload or responsibilities. Universities are large organisations in which relevant information may be held by academics, disability services, student services, accommodation teams and others. A defined institutional responsibility could provide greater clarity about who should receive information, when concerns should be escalated, who should act and where responsibility ultimately sits.
If a statutory duty required ever-increasing service provision without corresponding resources, the concerns about staff workloads and resources would have considerable force. Yet nothing considered here establishes that a statutory duty of care would necessarily take that form. Once again, this gets to the heart of the problem: different participants in this debate may be using the same term — “statutory duty of care” — while arguing for or against different things.
What Thompson's Argument Leaves Unanswered
In closing, it is worth returning to the title of the Westminster Hall debate: the potential merits of a statutory duty of care for universities. Thompson identifies a series of perceived disadvantages, drawing substantially on arguments advanced by UUK in 2023 and adding important UCU concerns about staff and resources. Any weighing of the pros and cons is largely absent from his contribution. But whether a statutory duty would provide an appropriate solution depends upon the problem it seeks to address and how the proposed duty is formulated.
The three situations considered earlier are intended as a simple way of explaining why greater precision matters. They distinguish between harm caused by what a university does, harm caused by its failure to do something for which it is already responsible, and harm arising from another source where a separate question of positive responsibility may arise. A statutory duty might concern the standard to which an existing responsibility is performed. It might address circumstances in which a positive responsibility arises. Or Parliament might define narrower duties relating to particular institutional responsibilities or particular categories of risk. Each formulation would raise different questions and have different merits, limitations and consequences.
One possible formulation illustrates the point. Where a higher education provider already has an acknowledged responsibility towards a student, should it be required to exercise reasonable care and skill in carrying out that responsibility? That is not presented here as the statutory duty Parliament should enact, nor is it the only possible formulation. But it is very different from imposing responsibility for preventing suicide, controlling autonomous adults, continually expanding student-support services or requiring universities to provide clinical interventions.
Greater clarity, consistency and individual accountability are at least potential merits that warrant consideration alongside legitimate concerns about resources, staff workloads, legal uncertainty and unintended consequences. Whether a particular statutory duty would address any of these problems depends upon what that duty is intended to achieve and how it is defined. That is the balancing exercise invited by a debate about potential merits.
Perhaps the most important conclusion is therefore a practical one. Annelli Peavot has argued for “formal forums where students, bereaved families, institutions, and government officials can discuss the gaps in duty of care”, describing this as essential for meaningful progress. Her suggestion deserves serious attention.
Discussion has too often involved an exchange of polarised positions rather than a meeting of minds. This issue has now been under sustained discussion for nearly four years. The launch of our petition at the end of 2022 brought it firmly into the public and political arena. Since then, there has been no shortage of argument. What has been missing is the kind of formal forum Peavot proposes, in which the different parties can examine those arguments together. Instead, positions have largely been advanced and answered through a prolonged sequence of petitions, government responses, parliamentary debates, briefings, correspondence and public commentary. Arguments are made, rejected and reformulated, often at different times and in different places, without the underlying questions necessarily being addressed.
Some of the most fundamental questions remain unresolved. What responsibilities do universities already have, and what standard should apply when those responsibilities are carried out? There are then further questions about when knowledge of a particular risk might require reasonable action, whether additional responsibilities should arise in particular circumstances, and how adult autonomy, institutional control, staff responsibilities and available resources should affect what can reasonably be expected.
These are difficult questions. They overlap, but they are not interchangeable, and many remain unanswered. Each requires careful examination if the potential merits and disadvantages of different formulations are to be properly understood. Without an opportunity to examine these questions in relation to one another, there is a real risk that different participants simply continue talking at cross-purposes. In its present form, that approach has taken us about as far as it can.
The next step need not be to secure immediate agreement on a particular formulation. A useful starting point would be more fundamental: to establish what responsibilities are already accepted, where the gaps lie, what additional responsibilities might reasonably be required and, ultimately, what everyone is actually arguing for — or against. Perhaps the time has come to bring the different parties together. A parliamentary roundtable would be a good place to start.