This article was originally published by Overland. Authors: Linda Briskman, Rita Jabri-Markwell, Associate Professor Mary Anne Kenny, Dr Victoria Mason, Sanam Mustafa and gwenaël velge.

On 3 September 2026, Australia’s university regulator, the Tertiary Education Quality and Standards Agency (TEQSA), gave universities just two weeks to respond to a proposal that potentially has far-reaching consequences for campus free speech and academic freedom. Its draft Statement of Regulatory Expectations requires universities to adopt a definition of antisemitism, with definitions of Islamophobia and racism against First Nations people to follow.

But what happens when a regulator committed to consistency asks universities to treat different forms of racism through different definitions?

In the space of a week, 380 scholars from thirty-four Australian universities, nearly half of them professors or associate professors, put their names to a submission by the Scholars for Academic Freedom and Due Process (SAFDP), a group formed in response to the consultation. Rather than asking universities to navigate separate and contested definitions for different groups, the submission proposes a common standard: whether the conduct dehumanises the people at whom it is directed on the basis of protected attributes.

When definitions create more questions

The first problem is the definitions themselves. TEQSA points universities towards two contested definitions of antisemitism: the International Holocaust Remembrance Alliance (IHRA) working definition and the Universities Australia (UA) definition. Both seek to distinguish antisemitism from legitimate criticism of the State of Israel, but both produce confusion about where the boundary lies. This matters because these definitions may inform complaints, disciplinary decisions and universities’ compliance with regulatory standards.

In August, NSW Supreme Court Justice Fagan described the IHRA core definition as “verbose and imprecise”, warning that its breadth risked “weakening the abhorrence that the word ‘antisemitic’ should properly evoke”. A year earlier, the Federal Court held that criticism of Zionism is criticism of “a philosophy or ideology“, rather than a racial or ethnic group. Kenneth Stern, the lead drafter of the IHRA definition, has said that the definition was never meant as a disciplinary tool, and that it was now routinely weaponised for political purposes.

The UA definition raises its own questions of clarity. Asked during a NSW parliamentary inquiry whether advocating a binational state with equal rights for all its citizens would fall within its reference to “eliminating the State of Israel”, the University of New South Wales said that it would not. Yet when repeatedly asked whether the wording should therefore be amended to make that clear, the university declined to say that it should.

These are not abstract problems. If a definition is to guide decisions about complaints and disciplinary action, students and staff need to be able to distinguish political or academic argument from discriminatory conduct with reasonable certainty.

A chilling effect

The apparent rationale for detailed definitions is that they help universities distinguish antisemitism from legitimate criticism of Israel.

There is growing evidence that uncertainty about where this boundary lies can affect academic expression. A Climate of Fear, surveying 527 respondents, and Erasure in the Academy, drawing on more than 150 submissions from over twenty campuses, reported a chilling effect on speech concerning Palestine and fears of institutional consequences. Similar concerns have been documented in Europe, Canada and the UK.

This is not new. In a 2017 legal opinion on the IHRA definition, barrister Hugh Tomlinson warned of a “chilling effect” where public bodies, faced with definitional uncertainty, might sanction conduct labelled as antisemitic “without applying any clear criterion of assessment”.

A hierarchy of racism

A second problem with TEQSA’s approach is consistency.

The new standard envisages a whole-of-institution approach to racism, yet the regulatory framework specifically identifies antisemitism, Islamophobia, and racism against First Nations people. Other forms of racism, including anti-Palestinian and anti-Arab, do not receive the same specific treatment.

This is surprising given that the Australian Human Rights Commission’s 2026 Racism at Uni study, drawing on more than 76,000 students and staff across forty-two universities, found high reported experiences of racism across multiple communities. Jewish and Palestinian respondents both report experiences of racism at rates over 90 per cent, while First Nations, Chinese, Middle Eastern and Northeast Asian respondents reported rates above 80 per cent.

This raises an important question: why should different forms of racism be governed through different definitional frameworks?

The concern is not that antisemitism, Islamophobia or racism against First Nations people should receive less protection. They require effective protection. The question is whether that protection is best achieved through a series of group-specific definitions, or through a common standard capable of recognising racism consistently, irrespective of the identity of its target.

That question becomes particularly important in universities, where difficult and sometimes deeply confronting political, historical and legal questions must be capable of being examined. An anti-racism framework must protect students and staff from discriminatory conduct and protect the ability to discuss, criticise and debate states, governments, political ideologies and the findings of international courts and human rights bodies.

If that is the case, why is TEQSA only addressing three types of racism? Does this approach not create a hierarchy of racisms? A hierarchy made starker juxtaposed with the International Court of Justice finding Israel in breach of the Convention on the Elimination of All Forms of Racial Discrimination, and the UN’s Commission of Inquiry concluding genocide had occurred in Gaza — findings that students and academics now risk being disciplined for repeating.

This is doubly worrying coming from institutions for whom empirical grounding and universal values have always been defining features.

The alternative

SAFDP proposes a different approach: a dehumanisation-based test, developed by the Australian Muslim Advocacy Network. Rather than creating separate regulatory tests for different groups, it applies a common standard to conduct directed at people because of protected attributes. AMAN argues that a single standard for judging complaints can reduce the risk of inconsistent outcomes. It is not intended to be a substitute for community-led processes to address structural and systemic racism on campus – it’s specifically for complaints about expression.

The framework draws on established hate-speech jurisprudence and research, including elements of the “hallmarks of hate” that the Supreme Court of Canada endorsed in Saskatchewan (Human Rights Commission) v Whatcott. It asks whether a group is portrayed, for example, as inherently criminal, dangerous or evil; collectively responsible for the conduct of some members; a uniform threat; or as vermin, filth or contaminants. Crucially, the framework protects people on the basis of protected attributes, not states, governments or political ideologies.

The framework seeks to identify language that crosses the line between legitimate speech, however forceful, and denying a group’s basic humanity. It does not adjudicate whose account of a conflict, an occupation, or historical event is correct. The standard does two things at once. It protects people from being treated as less than human, and it protects the speech needed to name abuse and to argue about states and ideologies. Both serve the same aim: maximising human rights, which belong to people rather than to states or ideologies.

There is also an important practical advantage. TEQSA itself says definitions should be conduct-focused, legally compatible, operationally usable, and accompanied by safeguards for academic freedom, lawful expression, procedural fairness and viewpoint neutrality. A common dehumanisation standard is designed to meet precisely those concerns.

With this framework, universities can adopt a common standard capable of applying across all forms of racism. In doing so they could go beyond TEQSA’s Regulatory Expectations by extending consistent protection to forms of racism not identified in the Regulatory Expectations.

Universal values and consistency should be the backbone of any legal or disciplinary standard: the same conduct should be assessed by the same principles, regardless of the identity of the group targeted. A clear conduct-based test may also reduce the legal and reputational risk associated with applying contested definitions whose scope and operation remain uncertain.

---

Dr Victoria Mason is a Senior Lecturer and Associate Dean of Inclusion and Equity at the School of Humanities, Arts and Social Science, Murdoch University.

Associate Professor Mary Anne Kenny teaches and researches in the area of human rights, refugee, and immigration law at Murdoch University's School of Law and Criminology, Murdoch University.

gwenaël velge is an Australian-based academic, research associate with the Mineral Policy Institute at Murdoch University.

This article is republished from Overland, with permission. Read the original original article.