Can Mediation Deliver Justice? Reflections on Bill Marsh’s UCL Lecture on Grenfell Tower, Bosnia and the Limits of Dispute Resolution
I recently attended a powerful and thought-provoking lecture hosted by the UCL Centre for Dispute Resolution, at which Bill Marsh addressed a deceptively simple but profoundly important question: can mediation deliver justice?
It was an important talk not because it offered an easy answer, but because it refused to do so.
In a field that can sometimes lapse into overstatement, Bill Marsh’s approach was careful, humane and properly nuanced. He did not argue that mediation is a substitute for courts, inquiries, criminal proceedings or public accountability. Nor did he suggest that mediators should pretend to wield some superior moral mechanism capable of curing every grievance. Instead, he explored a more difficult and more useful proposition: that although mediation cannot deliver justice in the same way that the formal justice system can, it may still have a vital role in helping people experience something that feels like justice.
That distinction matters.
Mediation now sits inside the “justice space”
One of the lecture’s starting points was historical. Bill reflected on the fact that when he began practising as a mediator some 35 years ago, mediation was seen as something separate from the civil justice system. Parties stepped out of litigation into mediation, and if it failed, they stepped back into the legal process.
That, of course, is no longer the position.
Mediation is now deeply embedded within the civil justice landscape. Developments over many years, culminating in the greater systemisation of mediation and reinforced by cases such as Churchill, mean that mediation is no longer merely adjacent to justice. In many contexts it is now woven into the machinery of civil dispute resolution itself.
That shift has consequences.
If mediation now operates within the “justice space”, then mediators and lawyers alike must confront a question that perhaps could once be avoided: what exactly are we offering when parties come seeking justice? Are we offering merely a pragmatic bargain? Are we offering a process that helps people move on? Are we offering some distinct form of justice? Or are we using the language of justice far too loosely?
Those are not abstract questions. They arise in real cases, and often in painful form.
As Bill observed, mediators meet people every week who say, in one way or another: “I want justice.” If that is what people seek, the profession cannot be content with vague or complacent answers.
Grenfell and Bosnia: two very different contexts, one common question
Bill approached the subject through two major examples from his own work.
The first was the Grenfell Tower litigation, arising out of the horrific fire on 14 June 2017 in which 72 people died, including one unborn baby, and many more were traumatised, bereaved, injured or made homeless. The resulting civil claims involved very large numbers of claimants and multiple institutional defendants, including those connected with cladding, fire safety, building ownership and contracting.
The second was a dialogue process in Zenica, a city in Bosnia and Herzegovina whose economy depends heavily on a major steel plant. The city, hemmed in by mountains, suffers severe pollution. The plant is economically indispensable, yet the environmental and health consequences are grave, with high levels of respiratory disease and cancer. The result is a community conflict involving suffering, dependence, grievance, distrust and questions of power.
These were obviously very different situations. One concerned a devastating urban fire and the aftermath of catastrophe in London. The other concerned industrial pollution, economic dependence and community conflict in Bosnia. But Bill identified several common denominators: a large number of people affected, deep suffering and trauma, and a passionate pursuit of justice by communities who felt gravely wronged.
That last point is central. In both contexts, justice was not a neat, singular concept. It was contested, emotional, personal and varied. For some, justice meant accountability. For some, punishment. For some, acknowledgement. For some, financial compensation. For some, institutional change. For some, being heard at last.
That diversity of meaning is itself one of the major difficulties in this area. It is also one reason why mediation can neither be dismissed too quickly nor romanticised too readily.
What does justice look like to those who seek it?
One of the most striking features of the lecture was Bill’s insistence that professionals working in and around dispute resolution should be slow to assume that they understand what people mean when they demand justice. Lawyers, judges, mediators and policymakers can easily fall into the habit of imagining that they already know the answer. But one of the lessons of Grenfell and Zenica was precisely that the voices of those affected must be heard in their own terms.
From those examples, Bill drew out several recurring expressions of what justice looked like to those involved.
“I need you to know”
The first expression of justice was the need for those responsible, or those perceived to be responsible, to understand the impact of what had happened.
In the Grenfell process, part of the settlement structure included what became known as Testimony Week: a public opportunity for claimants to speak directly, in the presence of senior representatives of the defendant bodies, about their experiences. This was not simply an evidential exercise. It was not merely a matter of proving loss. It was a forum in which those who had suffered could say, in effect: you must hear what this has done to us.
That is not something the law always does well. Courts determine facts, apply legal principles and award remedies, but they do not always create space for the full human reality of loss to be heard by those whose actions or omissions are said to have caused it. Mediation, or facilitation within a mediated or hybrid framework, may sometimes be able to create such a forum.
“I need you to change”
The second expression of justice was the demand for change. For many people, justice is not simply retrospective. It is not only about identifying what went wrong in the past. It is also about ensuring that the wrong is not repeated.
That was reflected, in the Grenfell context, in the desire that institutions and corporations go back to their boardrooms and fundamentally alter how they do business; that they place people ahead of profit; that the experience of Grenfell leads to real cultural and structural change.
This too is significant. Conventional legal remedies are not always well suited to meeting that need. Damages may compensate, but they do not necessarily transform institutional behaviour in a manner visible to those who have suffered.
“I need you to be accountable”
Accountability is an obvious component of justice, but the lecture illuminated its subtler dimensions.
Accountability is often thought of in formal legal terms: a finding of liability, a judgment, a conviction, a sanction. Those remain indispensable parts of the justice system. But Bill suggested that accountability can also take human and relational forms. The mere fact of turning up, listening, being present, and not hiding behind structures or lawyers can itself be meaningful.
That was one reason the attendance of senior representatives at Testimony Week mattered. It was not a complete substitute for legal accountability. It was not meant to be. But presence itself carried significance. Those who had suffered were no longer speaking into a void.
“What is your response?”
Perhaps the most difficult question of all was this: once the injured or bereaved have spoken, what response is given?
A genuine response cannot be compelled in the same way as a court order. It cannot simply be manufactured procedurally. If there is to be acknowledgement, remorse, responsibility or meaningful engagement, it must usually be freely given.
That, as Bill noted, is both mediation’s weakness and its strength.
It is a weakness because the forum can be created, but the quality of the response cannot be guaranteed. A party may attend, listen, and remain unmoved or at least unresponsive. Yet it is also mediation’s strength because where a response is freely given rather than extracted under compulsion, it may have a force and value of its own. Acknowledgement voluntarily offered can carry a moral weight that differs from an admission wrung out after defeat at trial.
In the Grenfell context, he referred to the response of the leader of the local council after Testimony Week as a remarkable example of leadership and acknowledgement. It helped, in his view, complete what he called a “circle of justice”: the need is expressed, and the response is genuinely given.
Bespoke outcomes and the shift in power
Two further themes were that justice often needs to be bespoke and that it often involves a shift in power.
Generic outcomes can feel impersonal. A legal system is bound, to a significant degree, to operate through categories, tariffs, remedies and doctrines of general application. That is necessary. But it can also leave individuals feeling unseen. Mediation has the potential, at least in some circumstances, to craft responses better suited to the actual needs of those involved.
In the Grenfell settlement, this included a restorative justice component and a fund directed not by external professionals imposing their own assumptions, but by the needs articulated by the affected community itself: education, trauma counselling, memorialisation and other projects shaped by those who had lived through the tragedy.
This connects to the idea of power. Many of those affected in Grenfell or in Zenica were people with little structural power. They were not accustomed to being heard, let alone to shaping the response. A mediated or facilitated process can, if designed carefully, alter that dynamic. It can create space in which those who were previously voiceless determine the structure, timing, content or direction of part of the process. That itself may contribute to a sense of justice.
Objective justice and subjective justice
The conceptual heart of the lecture, in my view, was Bill’s distinction between what he called, in effect, objective justice and subjective justice.
Objective justice is what people usually mean when they refer to the justice system in its formal sense. The state makes laws. Courts apply them. Conduct is judged against objective standards. Liability, innocence, guilt, rights and remedies are determined through recognised procedures.
Mediation cannot do that. It cannot substitute itself for adjudication, criminal sanction or the authoritative declaration of legal right.
Subjective justice, however, is different. It asks whether the person concerned feels that what happened was fair, or at least fair enough; whether they were heard; whether they were respected; whether they had some agency; whether anyone took responsibility; whether the outcome resonates with their lived sense of justice.
This is a useful distinction, though plainly not an absolute one. It avoids both exaggeration and dismissal. It does not claim that mediation can do everything. Nor does it relegate mediation to the narrow role of horse-trading over numbers. Instead, it recognises that there are dimensions of justice beyond formal adjudication, and that mediators may be able to contribute to those dimensions even where they cannot deliver objective justice itself.
For mediators, that is both an opportunity and a challenge.
A warning against overpromising
One of the most valuable aspects of the lecture was its clear warning against overclaiming for mediation.
There is, in some parts of the mediation world, a tendency to set up an unhelpful contrast: litigation bad, mediation good. Courts are caricatured as cold, combative and slow; mediation as humane, empowering and wise. That sort of rhetoric is tempting, but it is shallow and potentially irresponsible.
Litigation exists for a reason. Public adjudication, findings of fact, coercive powers and appeal structures matter. In some cases they are indispensable. Criminal accountability, in particular, cannot be replaced by carefully facilitated dialogue. Nor should it be.
If mediation is oversold as though it can provide all that courts provide and more, disappointment is inevitable. Worse, parties may feel misunderstood or even manipulated. A person who seeks public vindication, authoritative determination, precedent or punishment is not irrational. They are not failing to appreciate the virtues of compromise. They may simply be pursuing a different and entirely legitimate conception of justice.
This point is especially important now that mediation is increasingly woven into the civil justice system and, in some contexts, may become mandatory or strongly encouraged. The closer mediation sits to state-backed dispute resolution, the more carefully its limits must be acknowledged.
Mediation must not become too commoditised
A second warning was directed inwardly, at the mediation profession itself.
Bill expressed concern that mediation is becoming too commoditised: too standardised, too process-driven, too inclined to begin with a fixed model rather than with the actual needs of the parties. In simple, repeat-player commercial work, standard forms and predictable structures may have some utility. But in more complex, humanly charged or socially sensitive cases, an off-the-shelf model may be quite inadequate.
That criticism deserves attention.
There is always pressure in any professional field towards routinisation. Systems want models. Regulators want standardisation. Markets reward efficiency. None of that is inherently wrong. But if mediation becomes merely a packaged service with a familiar script, it risks losing precisely those features that make it capable of contributing to subjective justice: flexibility, responsiveness, creativity and attentiveness to the people actually involved.
The implication is not that all mediations must become sprawling bespoke social processes. Clearly they should not. Many disputes call for a straightforward, practical intervention. But where the issues are more profound, mediators must be willing to engage with complexity rather than forcing reality into a template.
Even the language matters
Another subtle but important point in the lecture was that what we call the process matters.
The word “mediation” does not mean the same thing to everyone. Different communities, cultures and participants may bring very different understandings to it. In some contexts, it may suggest imposed compromise. In others, gentle facilitation. In still others, shuttle negotiation or evaluative intervention.
That matters because language shapes expectation.
In Zenica, the process was described as a dialogue. In Grenfell, Bill and his fellow neutrals were referred to as facilitators. Those choices were not trivial. They reflected the fact that the participants’ understanding of the process, and their willingness to trust it, depended in part on how it was framed.
This is another salutary reminder that process design is not only structural but linguistic. If mediation is to contribute meaningfully in difficult cases, those who facilitate it must pay attention to how the process is presented, understood and owned by the participants.
The uneasy relationship with mandated mediation
Although it was not the principal focus of the lecture, the discussion touched on mandated mediation, and Bill made plain that he is not an enthusiast.
The objection was not doctrinaire. It was grounded in respect for party autonomy and moral seriousness. He gave the example of a party who was clear that what he wanted was a trial, and who had to incur the cost of mediation despite having no wish to compromise because doing so would, in his own terms, have betrayed his priorities and values.
That concern should not be brushed aside. If justice sometimes involves the right to pursue adjudication, then any movement towards compulsory mediation must be handled with caution. There may be strong arguments for encouraging early dispute resolution, and in many cases parties undoubtedly benefit from being required at least to engage. But there is also a danger that mediation becomes another procedural hurdle imposed upon those who, for principled reasons, seek something else.
For those of us committed to mediation, this is worth reflecting upon. Advocacy for mediation is strongest when it remains candid about what mediation can and cannot do.
What does all this mean for mediators?
The lecture did not provide a single model or formula. Indeed, it expressly resisted that temptation. But it did point towards a number of conclusions.
First, mediators need to be more thoughtful about justice as a real concern of real parties, not merely as a slogan.
Secondly, mediators should not assume that the only things parties want are money, certainty and risk management. Often they do want those things. But often they also want voice, acknowledgement, accountability, explanation, dignity or change.
Thirdly, mediators should design or adapt process with those needs in mind. That may involve different forms of encounter, opportunities to speak without interruption, careful sequencing, bespoke restorative elements, or hybrid approaches that operate alongside rather than instead of formal justice mechanisms.
Fourthly, mediators need humility. They are not saviours, and mediation is not a universal solvent. But humility should not mean timidity. Properly understood, mediation can sometimes create spaces that courts cannot.
Finally, the mediation community should continue to grapple with how its work intersects with the broader justice system. This is not merely a technical or procedural issue. It is a question of legal culture, public legitimacy and professional honesty.
A final reflection
My own impression on leaving the lecture was that Bill Marsh had performed a valuable service by asking the question in a way that resisted slogans.
Can mediation deliver justice?
If the question means, can mediation replicate adjudication, determine public rights, establish legal truth, punish wrongdoing and vindicate the rule of law in the way a court can, then plainly the answer is no.
But if the question is whether mediation, facilitation or structured dialogue can sometimes help people experience recognition, accountability, agency, acknowledgement and fairness in ways that matter deeply to them, then the answer is plainly yes.
The challenge is not to confuse those two propositions.
Mediation should never pretend to be the whole of justice. But nor should it undervalue its capacity to contribute to justice, especially where what people most need is not simply a legal answer, but a human one.
For those of us involved in dispute resolution, that is not a reason for complacency. It is a reason for sharper thinking, greater humility and more careful practice.
And perhaps that was the lecture’s most important lesson: mediation does not become more valuable by claiming to do everything. It becomes more valuable when it understands, with precision and honesty, what it can uniquely offer.