On the IHRA Working Definition

The IHRA working definition is thirty-eight words. In full, it reads:

‘Antisemitism is a certain perception of Jews, which may be expressed as hatred toward Jews. Rhetorical and physical manifestations of antisemitism are directed toward Jewish or non-Jewish individuals and/or their property, toward Jewish community institutions and religious facilities.’

Its own authors describe it as ‘non-legally binding’. It uses unclear language – ‘a certain perception’, and ‘may be expressed’ – that is vague where a working definition needs to be clear and precise. The second sentence is so broad that it adds little: it describes that antisemitism could have any form and be directed anywhere and at anyone, but not what makes something antisemitic. Legal experts who have examined it closely, including senior counsel and judges (Robertson, Tomlinson, Sedley and Fagan) have concluded that it does not do what a definition must do: it does not reliably distinguish what is antisemitic from what is not.

This matters most at the point where organisations actually use it. A definition applied in a complaints process, a disciplinary hearing, a funding decision or a code of conduct needs to give clear, consistent answers. The IHRA working definition does not. Different people applying it in good faith reach different conclusions about the same conduct, which is the opposite of what an organisation adopting it is hoping for.

The working definition is short, but it travels with a list of eleven illustrative examples, and seven of those eleven concern the state of Israel rather than hostility to Jewish people as Jewish people. One, for instance, treats the claim that the state of Israel is ‘a racist endeavour’ as antisemitic – placing a contested political and historical argument outside acceptable discussion (that framing, and others like it, are examined in the guide to contested language). Those seven are set out in full below. They are:

• ‘Accusing the Jews as a people, or Israel as a state, of inventing or exaggerating the Holocaust.’
• ‘Accusing Jewish citizens of being more loyal to Israel, or to the alleged priorities of Jews worldwide, than to the interests of their own nations.’
• ‘Denying the Jewish people their right to self-determination, e.g., by claiming that the existence of a State of Israel is a racist endeavor.’
• ‘Applying double standards by requiring of [Israel] a behavior not expected or demanded of any other democratic nation.’
• ‘Using the symbols and images associated with classic antisemitism (e.g., claims of Jews killing Jesus or blood libel) to characterize Israel or Israelis.’
• ‘Drawing comparisons of contemporary Israeli policy to that of the Nazis.’
• ‘Holding Jews collectively responsible for actions of the state of Israel.’

Three of these – collective responsibility, dual loyalty, and the charge of inventing or exaggerating the Holocaust – describe conduct that is antisemitic on any sensible test, and this guide treats it as such. The other four do not ask who was targeted or why. They identify a position, or a comparison, or a distribution of attention, and classify it as racism by definition. That is the source of most of the difficulty organisations encounter when they try to apply the definition in practice.

The status of these examples is itself confused: they were not adopted as part of the working definition by the IHRA decision-making body, and they were not adopted by the Australian Government. However, they are routinely applied as though they had been. For an organisation, this is the practical trap. You adopt what looks like a short, reasonable statement, and you inherit a list of examples that reach well beyond antisemitism into ordinary political debate about Israel – debate your staff, students or members are entitled to have. It is also the reason the examples deserve to be read in full before adoption rather than after. An organisation that has read them is unlikely to conclude that it has adopted only a short, reasonable and clear definition of antisemitism.

The clearest evidence of a definition’s effect is how it has actually been used. The record of the IHRA working definition, in Australia and overseas, is that it is used more often to restrict criticism of Israel than to protect Jewish people from antisemitism.

The European Legal Support Centre reviewed dozens of cases in which the definition was invoked against academics, students, journalists and artists. In a study of forty university cases, not one allegation was ultimately upheld. Those accused still endured months of investigation and reputational harm, and others fell silent rather than risk the same. The people most often affected have been Palestinian people speaking about their own history, and Jewish people who dissent from established positions on Israel. This is not a marginal side effect. It is the definition’s principal documented use. Similar outcomes are documented in Australia.

An organisation weighing this decision is not alone, and adopting the definition is not the cautious choice that will minimise risk. In May 2026 the ABC and SBS both declined to adopt the IHRA working definition, citing its ambiguity, and chose to rely on their own clearer guidelines. Several Australian universities, including the ANU, declined to adopt it. Internationally, more than a hundred civil society and human rights organisations urged the UN not to adopt it, and UN human rights experts have recommended that states pause or rescind their adoption of it, finding it inconsistent with international human rights law. In Australia Jewish groups supported by other civil society organisations have urged the government not to use it.

The definition’s own lead drafter, Kenneth Stern, has said repeatedly that it was never intended to be used as a speech code or a disciplinary instrument, and has warned against exactly the institutional uses it is now being put to.

An organisation adopts a definition of antisemitism to reduce confusion, protect Jewish members and act consistently. The IHRA working definition tends to produce the opposite outcomes. Because it is contested, adopting it draws the organisation into the political dispute it was trying to avoid. Because it is imprecise, it produces inconsistent decisions that are difficult to defend. And because its best-known use is the restriction of Palestinian and dissenting-Jewish expression, adopting it can divide the community it was meant to reassure – including the Jewish community, which does not speak with one voice on this question.

There is a better approach, and it is one an organisation can apply with more confidence and less risk. We set it out below.

In Executive Council of Australian Jewry v Haddad [2025] FCA 720, the Federal Court of Australia considered whether strong political criticism of Israel amounted to unlawful racial vilification under the Racial Discrimination Act. Justice Stewart held that it did not, finding that ‘the ordinary, reasonable listener would understand that not all Jews are Zionists or support the actions of Israel,’ and that ‘disparagement of Zionism constitutes disparagement of a philosophy or ideology and not a race or ethnic group.’ Political criticism of Israel, the Court said, ‘however inflammatory or adversarial, is not by its nature criticism of Jews in general.’

The practical consequence for an organisation is direct. The IHRA working definition’s examples treat much of this same criticism as antisemitic. An organisation may adopt the definition and then act on it, by disciplining a staff member, penalising a student, refusing a venue or withdrawing funding because someone criticised Israel or Zionism. In doing so it risks treating as misconduct the expression an Australian court has found to be lawful. That exposes the organisation to a challenge by the person it acted against – in an employment case, for example, through an unfair dismissal claim.

In State of New South Wales v Farhat [2026] NSWSC 985, Justice Fagan of the NSW Supreme Court examined the IHRA working definition’s formulation closely and rejected it as a guide to the meaning of antisemitism. He described the wording as ‘verbose and imprecise’ and ‘detached from the reality of the common usage and understanding of antisemitism in this country,’ and found that it ‘appears to be an attempt to repurpose the term “antisemitism” as the name of an amalgamated concept of “hatred toward Jews” and/or criticism of Israel’. He confirmed that ‘As an ideology of Jewish nationalism, Zionism is open to political discussion and criticism, notably on grounds concerning the rights of the Palestinians.’

Justice Fagan also addressed the argument organisations most often rely on – that because governments and institutions have adopted the IHRA working definition wording, it now defines the term. He rejected it. The fact that governments and institutions had adopted the formulation did not, he held, ‘change the established meaning of a word that has been in use in the English language for 150 years.’ Adoption by policy does not convert a contested definition into the legal meaning of antisemitism.

An organisation weighing adoption should understand three things:

• The IHRA working definition does not carry the legal authority its promoters suggest.
• Australian courts have drawn a clear line between antisemitism and criticism of Israel or Zionism.
• Acting on the IHRA working definition in the way it is designed to be used can place the organisation on the wrong side of that line.

Adoption of the working definition may therefore not reduce an organisation’s legal exposure, but instead may introduce exposure it had not anticipated. An organisation weighing adoption should obtain legal advice.

Australian law already contains a clear, workable test for racism, in section 18C of the Racial Discrimination Act 1975. The question it asks is simple: was an act reasonably likely to offend, insult, humiliate or intimidate another person or a group of people because of their race, colour or national or ethnic origin? The words that matter most are ‘because of’. Offence alone is not enough. The act must be done because of the person’s race, colour or national or ethnic origin. Applied to antisemitism, the question is equally simple: was a person, or a Jewish institution, targeted because they are Jewish?

This is the test we set out in full in our guide to recognising antisemitism, and it has three advantages over the IHRA working definition for any organisation. It is clear, so that different people applying it reach the same answer. It is consistent, because it is the same test used for racism against every other community, which means an organisation is not creating a separate and unequal process for one group. And it is already law, so it rests on decades of established meaning rather than a contested definition. Australian courts have shown they can apply exactly this test – distinguishing genuine antisemitic vilification, which is unlawful, from political criticism of Israel or Zionism, which is not.

Antisemitism has distinctive features – a long history of conspiracy claims, the shadow of the Holocaust, and the persistence of particular stereotypes – and responding to it requires understanding them. But antisemitism is racism: the targeting of people for who they are. The strongest, best-evidenced response treats it as part of the shared response to all racism, rather than as a separate category with its own definition, laws and institutions.

In November 2024 the Australian Human Rights Commission delivered the National Anti-Racism Framework, developed through consultation with communities across the country, including Jewish communities, and drawing on the experiences of more than 1,200 people with lived experience of racism. It offers organisations a ready-made, evidence-based approach that is human-rights-based, evidence-led, community-centred and whole-of-society. An organisation looking for a credible way to address antisemitism does not need to adopt a contested new definition; it can align with the framework Australia’s own human rights body has already built.

The National Anti-Racism Framework approach reflects the consensus of contemporary anti-racism practice:

• It is based on human rights, so that protecting one community does not come at the expense of another community’s rights.
• It is evidence-based, relying on consistent and transparent data about the racism that is occurring and about what responses are effective.
• It is community-centred, involving the people affected by racism, in their full diversity, in the design of the response.
• It is whole-of-society, engaging education, media, workplaces and government in addressing the conditions that produce racism, rather than relying on the criminal law as the primary response.

The approach we set out applies that framework to antisemitism: define antisemitism clearly, measure it accurately, involve the full diversity of Jewish Australians, protect legitimate political expression while addressing genuine hostility, and build the connections between communities that make every community safer.

Treating antisemitism as separate from, more serious, or as a bigger problem than other forms of racism – with a broader definition, distinct laws and separate institutions – is often proposed in good faith, as a way of recognising the seriousness of antisemitism. The evidence, however, indicates that it produces three effects that work against addressing anti-Jewish racism.

It can reduce solidarity between communities. This is especially true where one community’s protection appears to come at the cost of another community’s ability to speak about its own history. Research describes this pattern as competitive grievance: communities are set against one another, and hostility increases toward the community seen to receive more favourable treatment.

It can separate the response to antisemitism from its natural allies. The response to racism in Australia has been led by the communities that experience it most directly, First Nations peoples above all, and each community’s safety has been advanced through shared effort. An approach that places the response to antisemitism in opposition to the response to anti-Palestinian racism, or to legitimate expression, separates Jewish Australians from those alliances.

It does not address the main source of antisemitism. In Australia, the evidence shows, serious antisemitism comes overwhelmingly from far-right politics, which also targets Muslim and Arab Australians, migrants and First Nations peoples. An approach that directs institutional attention toward political speech directs it away from the far right.

• Apply the same test to antisemitism as to other forms of racism: is the person targeted because of their identity – that is, in the case of antisemitism, because they are Jewish? Australia’s existing racial discrimination law asks this question, and the courts have shown that it can be applied, finding genuine antisemitic vilification unlawful while protecting political expression about Israel and about Zionism.
• Strengthen protections for all communities together, through consistent anti-discrimination and anti-vilification law, rather than through separate arrangements for some communities.
• Educate about antisemitism accurately in the context of all forms of racism – its history and its present-day forms – alongside education about the racism experienced by other Australian communities.
• Collect data carefully and transparently so that public understanding rests on evidence that communities can trust.
• Support solidarity between communities. The safety of Jewish Australians is connected to the safety of every community that experiences racism, and the communities that experience racism are, in practice, one another’s most effective defenders.