The Miller judgment draws its line through students

The Employment Appeal Tribunal has rejected the University of Bristol’s challenge to the finding that it discriminated against David Miller when it sacked him in 2021.

Miller, a professor of political sociology, was summarily dismissed for gross misconduct in October 2021 over three sets of comments made in February that year.

There was a speech at a “Building the Campaign for Free Speech” event, comments to the Jewish Chronicle, and an on-the-record email to a student journalist – in which he described Zionism as “a racist, violent, imperialist ideology premised on ethnic cleansing” and characterised Bristol’s Jewish Society and the Union of Jewish Students as Israel lobby groups using Jewish students as “political pawns.”

The tribunal found his anti-Zionist beliefs protected, his dismissal directly discriminatory, unfair and wrongful – but also found what he said about students and student societies blameworthy enough to cut his unfair dismissal compensation by 50 per cent.

Counsel for Bristol threw nine separate arguments at the tribunal’s decision, most with multiple sub-arguments underneath them and some with sub-sub-arguments. Eight of them failed. Miller ran his own appeal on three points, and lost all three.

The one point Bristol did win concerns how much compensation Miller might eventually get – and even that has been sent back to the same tribunal to have another think.

The judgment was handed down by a panel chaired by the EAT’s President Lord Fairley, following a hearing last November.

When the employment tribunal found for Miller back in February 2024, I noted on the site that an ET judgment doesn’t set legal precedent in and of itself, however closely it gets studied. That caveat has now expired.

This is an appeal decision, which means employment tribunals across Great Britain now have to follow it. And what it settles is that the specific anti-Zionist beliefs Miller set out in his claim – that political Zionism, as he defined it, is inherently racist, imperialistic and colonial and ought to be opposed – are protected beliefs under section 10 of the Equality Act 2010.

Protection is the status the Act gives to religion or belief. Once a belief qualifies, an employer can no more sack you for holding it than for your race or your religion. That doesn’t hand every future anti-Zionist claimant a win, mind – it mattered here that the tribunal found Miller sincerely held those beliefs, and that he wasn’t supportive of violence as a means of opposing Zionism.

Both sides appealed, and the result splits neatly in two. One half sets out what a university can’t touch – the belief itself, and expressing it. The other half sets out what it still can – what an academic says in public about its own students. Everything after that is about the size of the eventual cheque.

The concession that decided it

The first question in a case like this is whether the belief qualifies for protection at all. If it doesn’t, nothing else matters, because there’s nothing for discrimination law to bite on.
So the lead argument on appeal for Bristol was that the tribunal had “impermissibly stripped away inconvenient aspects of C’s true beliefs to focus on its core” – or in plainer terms, that it had protected a tidied-up version of what Miller actually thinks.

His real views, counsel submitted, extended to an openness to violence as a means of opposing Zionism. Had the tribunal looked at the beliefs “in their totality”, they’d have flunked the Grainger test – the checklist courts use to decide whether a belief is sincerely held, weighty, coherent and worthy of respect in a democratic society. Fail it, and the belief gets no protection at all.
The difficulty was a transcript. On day one of the tribunal hearing in October 2023, Bristol’s own counsel had told the panel:

I made it clear at the previous preliminary hearing and I make it clear now, I do not seek to suggest an alternative belief… If I can be absolutely clear, I am not seeking to go behind what the claimant says his true belief is.

You can’t promise a tribunal you won’t run an argument and then run it on appeal. The EAT calls that concession “clear and unequivocal”, notes that no Bristol witness gave evidence of any different “true belief” either, and concludes the university can’t now go back on it. One passage of transcript sees off the main ground of appeal.

And in an aside that will be useful to anyone preparing for the next case of this kind, the EAT suggests it may not have been proper to put those questions to Miller in the first place – the “relevance and propriety” of asking him whether he was “open to” violence, it says, “may be questionable.”

The EAT works through the substance anyway, and three findings are worth having. Political beliefs, as well as religious ones, can qualify for protection – Bristol’s counsel abandoned his written argument to the contrary once he got to his feet, “correctly in our view.”

A belief that grows out of academic research is still a belief, rather than just a position held on the current state of the evidence. If it were otherwise, every academic’s convictions would fall outside the Equality Act precisely because they’d been researched.

And on coherence – one of the Grainger conditions, and the one counsel for Bristol said Miller’s beliefs failed – the EAT goes further than it strictly needed to:

It is coherent to describe as ‘racist’ an ideology that promotes the establishment of a state for only one race of people in a territory that contains a large number of people of a different race. Such an ideology, which supports the migration of members of the first group into the territory with the support of an imperial power to dislodge an indigenous population, could also coherently be described as colonial and imperialistic.

Coherent is not the same as correct, and whether the belief is right or wrong, the EAT stresses, was never the question. But that paragraph – an appeal tribunal spelling out why “Zionism is racist” is a position that hangs together well enough to attract legal protection – I suspect will end up being quoted for years, by people on every side of this argument.

One bad reason spoils the lot

The second question is why the university acted. This is the part that decides whether it broke the law.

Bristol’s problem is that it didn’t have one reason for sacking Miller. The EAT reads the tribunal as having found two, bundled together.

The first was Miller simply expressing his anti-Zionist beliefs – to which, on the tribunal’s findings, no reasonable objection could be taken. The second was his comments aimed at students and student societies, which were blameworthy, and which could be pulled apart from the belief itself.

That distinction is the heart of the case, and it comes from two earlier Court of Appeal decisions, Page and Higgs. The rule they set up is that an employer can sometimes act over the objectionable way a belief gets expressed – as long as what it does is a proportionate response to the manner of the expression, rather than to the belief.

But where part of the reason for acting is the belief itself, or an expression of it that nobody could reasonably object to, there is nothing left to weigh up. It’s discrimination, and the question of whether the employer’s response was proportionate never arises.

And because a real part of Bristol’s reason was, in the EAT’s words, “simply his non-objectionable expression of his protected beliefs,” that’s exactly where this dismissal landed:

…the application of Higgs did not arise, and the issue of justification was not one that required to be determined by the Tribunal.

The cross-examination of Professor Norman, the Dean who fired Miller, showed how that worked in practice. She accepted that if he’d linked student groups to a cause in a pro-Zionist context – noting that certain student societies were constitutionally bound to promote certain interests – she wouldn’t have treated it as gross misconduct.

Asked what her decision would have been if the references to Zionism being racist had been stripped out of the February comments, she replied: “that did not happen.”

So the rule for employers is this. If the protected belief, or an unobjectionable expression of it, forms a real part of why you acted, the decision is discriminatory and no amount of argument about proportionality will rescue it. Conduct that genuinely stands apart from the belief can still be disciplined. Which means the reasoning a panel writes down matters enormously – one contaminated strand is enough to sink the lot.

The EAT then runs the proportionality analysis anyway, in case it’s wrong about that, and upholds the tribunal on every point, endorsing this passage in particular:

The University, as an academic institution, ought to be prepared to face and to weather criticism and reputational damage which flows from the exercise by its academics of their rights to speak and think freely and lawfully on areas within or connected to their research and expertise.

Part of what made dismissal disproportionate was consistency – or the lack of it. No disciplinary action had been taken against another Bristol academic, Professor Greer, over comments he’d made in the Daily Mail about a student society in late 2021, and the tribunal was entitled to weigh the difference in treatment. Counsel objected to that too, but got nowhere.

The seven headline criticisms and thirty sub-points on proportionality advanced for Bristol get described as “no more than expressions of disagreement with the conclusion reached by the Tribunal and an attempt to re-try the issue.”

Its counsel, we also learn, used his entire agreed time allocation on his first speech and left himself nothing for a reply. You can feel where the judicial patience wore thin.

The line runs through students

Miller lost his own appeal in full – and this is the half universities will want to read twice, because it’s where the EAT sets out what an employer can still do.

The finding that he committed misconduct stands, because his comments about JSoc, UJS and identifiable students weren’t an expression of the belief he’d set out in his claim.

That belief, as both sides agreed at the outset, was that political Zionism is inherently racist, imperialistic and colonial and ought to be opposed. It said nothing about students or student societies. Hence the line the EAT quotes back approvingly from the tribunal:

It was clearly open to the claimant to articulate his views about Zionism without reference to students and student societies.

So the misconduct finding survives. So does the dismissal of his separate claim that the university’s investigating officer harassed him. And so does the 50 per cent cut to his unfair dismissal compensation – tribunals can reduce an award where the employee’s own conduct contributed to what happened, and here the tribunal held employer and employee “equally to blame”.

The EAT also upheld the finding that treating the comments about students and student societies as misconduct was compatible with Article 10 – the right to free expression, which universities are bound to respect.

Bristol’s Free Speech Code gave it the clear published basis that any restriction on speech needs, and the power imbalance between a professor and students, along with the risk of reputational harm, could justify restricting what he said about them. What none of that justified was jumping straight to dismissal.

Put the two halves together and you get something close to a workable rule. A university can’t discipline an academic for holding or expressing the view that Zionism is racist and should be opposed – however much correspondence arrives demanding that it does.

It may be able to discipline the same academic for aiming that at its own students and their societies in public. But the sanction has to be proportionate – here that meant something short of dismissal, with a warning expressly identified as available – and the university had better be able to show that the belief itself, or an unobjectionable expression of it, formed no part of why it acted.

Miller was sacked on the spot for what Bristol called gross misconduct – and no tribunal, at either level, has found that anything he did amounted to gross misconduct. That’s why the wrongful dismissal claim succeeded and survived appeal. Wrongful dismissal is a contract claim rather than a discrimination one, and the point is simple enough – his conduct wasn’t serious enough to tear up the contract, so sacking him without notice was itself a breach of it, whatever else it was.

And the late attempt on Bristol’s behalf at a different route – arguing that Miller’s comments amounted to unlawful victimisation of the students who’d complained about him, which in Equality Act terms means treating someone badly because they’ve complained about discrimination – was shut out on procedural grounds.

It had been expressly abandoned at an earlier case management stage, never written into Bristol’s case, and never put to Miller when he gave evidence. Whether an academic publicly targeting students who’ve complained about him is itself an Equality Act breach that an employer could rely on is a question this judgment leaves entirely open. Someone may run it properly next time.

One trap left open, mind. The whole students-are-separable analysis works because the belief Miller set out in his claim never mentioned students. The EAT notes that if it had, “difficult questions might potentially have arisen” about whether comments about them could have qualified for protection at all – questions that simply weren’t in front of the tribunal.

The next claimant’s solicitors will have read that paragraph too, and may try to define the belief more widely, so that comments about students sit inside it rather than outside. But a wider definition would run straight into the difficulties the EAT flagged and left unanswered. Whether students stay separable under a broader claim is the question this judgment doesn’t reach.

30 per cent

Everything so far is about liability – who was in the wrong. Bristol’s solitary win is about the bill.

When a tribunal works out compensation, it can discount the award to reflect the chance that the employee would have lost the job anyway. Here it put that chance at 30 per cent, on the basis that Miller might have been fairly dismissed within two months of his August 2023 tweets – the ones asserting that “Judeophobia barely exists these days”.

The EAT’s response:

…we do not understand why the Tribunal reached the conclusion it did.

It was never suggested at trial on Bristol’s behalf that those tweets were antisemitic, Professor Norman was “equivocal” about them when she gave evidence, and nobody analysed whether they could lawfully have grounded a dismissal at all. That finding is set aside and sent back to the same tribunal to deal with at the compensation hearing.

It cuts both ways. The objection is the absence of any explained basis for a chance of lawful dismissal, or for putting it at 30 per cent – so on reconsideration the figure could land lower as easily as higher. Compensation for the discrimination claims was left to that hearing anyway, so the money – nearly five years after the dismissal – remains entirely to play for.

Bristol also remains free to argue there that the discrimination award should be cut in the same way as the unfair dismissal one. The EAT’s point was that the question hasn’t been decided yet, not that it can’t be.

Where it goes from here

IHRA – the working definition of antisemitism that consumed most of the oxygen when this row was live in 2021 – played no part in the issues the EAT had to decide. Bristol conceded back in 2023 that it didn’t dismiss Miller for antisemitism, and that its position matched the two McColgan KC reports that found no case to answer.

Counsel for Bristol also formally argued that Higgs itself was wrongly decided. That argument had nowhere to go at the EAT, which is bound by Court of Appeal decisions, and it was advanced purely to keep the point alive higher up – which is a hint that the Court of Appeal is at least being contemplated.

But even winning that wouldn’t necessarily remove the discrimination finding. The EAT’s main conclusion was that the Higgs framework never came into play here, because an unobjectionable expression of

Miller’s belief was itself part of why Bristol dismissed him. Everything the EAT said about proportionality was a backup in case that was wrong.

And discrimination isn’t the only obstacle. The tribunal separately found that dismissal fell outside the range of responses open to a reasonable employer, and that Miller’s conduct didn’t justify sacking him without notice. Both of those findings have survived the appeal as well.

The lesson for the sector is a sequence rather than a slogan. Separate the protected belief from any misconduct that genuinely stands apart from it, identify the policy you’re relying on to intervene, evidence the harm or risk you say has been caused, treat comparable cases consistently, and think properly about sanctions short of dismissal.

The volume of complaints and the weight of reputational pressure make that job harder, but they don’t change the fundamentals.
_______________________________________________
A spokesperson for the University of Bristol said:

We recognise the Employment Appeal Tribunal’s judgment, although we are disappointed with its findings.
Our core objectives are to ensure all our students thrive at the University of Bristol and to protect freedom of speech and expression, a fundamental part of that care and consideration. At the same time, we have an obligation to make sure that these essential and important freedoms are exercised appropriately and in line with our codes of conduct and the expectation of the highest standards of behaviour.
We will continue to provide an environment where our staff and students can debate difficult and contentious issues, fostering a positive working and learning environment where free speech and academic freedom are championed.