The JDA Definition of Antisemitism is Completely Incoherent and Much Worse than the IHRA Definition – Ari Allyn-Feuer


Despite over fifteen years of controversy about the IHRA (formerly EUMC) definition of antisemitism[1], and despite large numbers of people passionately insisting that the IHRA definition is fatally flawed and unfit for purpose, and we simply must have a better definition, they haven’t actually produced an alternative definition with any real support or staying power.  The Jeremy Corbyn-era British Labour party had a few fiascos when they tried to make up their own definitions and adopt them; first a completely homebrew definition[2], and then a truncated version of the IHRA definition which omitted examples of antisemitic antizionism[3].  Both of these melted down immediately when the definitions were revealed, and it became clear they were merely slapdash attempts to legitimize Labour antisemitism by defining it out of existence. 

Since then, IHRA critics have seemed much more eager to make sweeping statements about what is not antisemitic than to propagate a definition of what is.  We are frequently treated to statements like “antizionism is not antisemitism,” or even “antizionism cannot be antisemitic,” and declarations that members of the left, being anti-racist, therefore cannot be antisemitic or engage in antisemitism.  Antisemitism is frequently, explicitly or implicitly, reduced to a matter of intent, with the possibility of inadvertent and institutional antisemitism wholly dismissed.  But there was never a concerted attempt to write down, circulate, and build consensus for an alternate positive definition of antisemitism. 

This gaping absence understandably fostered a suspicion among anti-antisemites that the criticism of the IHRA definition was driven not by a desire to formulate a definition which would permit society to recognize and combat antisemitism, but by a desire to legitimize acts deemed antisemitic or suspect by the IHRA definition, with at best indifference to the consequences for anti-antisemitism and its effort to protect Jews.  This current of thought is summarized by Lars Fischer[4] in his declaration that “Only Antisemites Oppose the IHRA Definition of Antisemitism.”

Until now, that is.  Now, the Jerusalem Declaration on Antisemitism (JDA) has emerged[5], and as its centerpiece published a definition of antisemitism.  While matters are still early, and it is quite possible that a few years from now this JDA definition will be as defunct socially as Labour’s abortive home-made definition from a few years ago, it is certainly true that this new definition has attracted much more support and comment than prior such definitions.  It has signatures from over two hundred academics including many scholars of antisemitism.  And its signatories and supporters have organized a formidable PR blitz, which has resulted in positive coverage in, among other places, Slate[6], Inside Higher Education[7], the Middle East Eye[8], the EU Observer[9], Jewish Currents[10], and the Forward[11].  Along with this, the new definition has been endorsed by groups like Jewish Voice for Peace and the BDS Movement (more on this later).

IHRA opponents did, at long last, what many anti-antisemites had been asking them to do for over fifteen years.  I take this occasion to subject the new proposed definition to a careful examination, both of the context and history, the text itself as versus the IHRA, and its reception.  I find that it is lacking in any merit other than the capacity to conceal antisemitism.

The argument of JDA definition proponents

Advocates of the new JDA definition say that it should be adopted “as an alternative to the IHRA Definition” because it is a better definition; they say both that the IHRA definition is bad, and that the JDA definition is good. 

In the declaration and its FAQ, they say that:

  • “Because the IHRA Definition is unclear in key respects and widely open to different interpretations, it has caused confusion and generated controversy, hence weakening the fight against antisemitism.”
  • The IHRA definition “puts undue emphasis on one arena” because seven of its eleven examples are about Israel, Palestine, and zionism.
  • The IHRA definition “is neither clear nor coherent…blurs the difference between antisemitic speech and legitimate criticism of Israel and Zionism…causes confusion, while delegitimizing the voices of Palestinians and others, including Jews, who hold views that are sharply critical of Israel and Zionism.”

That is, they identify four key problems with the IHRA which they wish to solve.  They wish their new definition:

  1. To be clear and coherent.
  2. To clearly separate antisemitic speech from legitimate criticism of Israel.
  3. To avoid confusion and controversy, and build consensus and trust.
  4. Thereby to help the fight against antisemitism.

And they are quite, quite certain that they have!  The JDA declaration describes its definition like this:

  • “The JDA reflects the clear and authoritative voice of scholarly experts in relevant fields.”
  • “The JDA benefits from several years of reflection on, and critical assessment of, the IHRA Definition. As a result, it is clearer, more coherent and more nuanced.”
  • “The JDA articulates not only what antisemitism is but also, in the context of Israel and Palestine, what, on the face of it, it is not. This is guidance that is widely needed. “
  • “The JDA invokes universal principles and, unlike the IHRA Definition, clearly links the fight against antisemitism with the fight against other forms of bigotry and discrimination.”
  • “The JDA helps create a space for frank and respectful discussion of difficult issues, including the vexed question of the political future for all inhabitants of Israel and Palestine.”
  • “For all these reasons, the JDA is more cogent, and, instead of generating division, it aims at uniting all forces in the broadest possible fight against antisemitism.”

Similarities between the two definitions

It is useful and important to note the many similarities between the two definitions.

  1. Both definitions declare that their intent is to damp controversy, and coalesce consensus and clarity around a durable definition, which can be used to recognize and fight antisemitism.
  2. Both definitions are composed of a very brief positive definition, and a longer commentary centered on both positive and negative examples.
  3. Both definitions address the bulk of their examples to Israel, Palestine, and zionism.
  4. Both definitions specify explicitly that advocacy of human rights and for Palestinian liberation and statehood are not inherently antisemitic, nor is criticism of Israeli government policy, or of zionism.
  5. Both definitions recognize that some Palestinian advocacy, or Israel criticism, is antisemitic, and that some antisemitism veils itself in these legitimate activities.
  6. Both definitions specify explicitly that context and judgment are central to the application of the examples.
  7. Both definitions are not legally binding and do not propose to ban speech or activity except as otherwise provided by law; they only provide the opportunity to judge whether speech or activity is antisemitic, not to proscribe it legally.

The argument of IHRA definition proponents

In the reaction to the JDA definition by advocates of the IHRA definition like David Hirsh[12], Dave Rich[13], Lars Fischer[14], Ben Cohen[15] and David Collier[16], a common theme has been the insistence that the IHRA definition is, as David Hirsh puts it[17], “a material social fact.”  They point out that the IHRA definition emerged from the necessity of a definition that addressed the prevalent and threatening phenomenon of Israel-focused antisemitism[18].  It has been widely adopted and used by institutions around the world, including the EU, OAS, and UN, dozens of countries, and hundreds of universities.  These institutions rely on it for ongoing use in the urgent work of combating the antisemitism that really exists in their communities.  And Jews around the world rely on it to guide their communities in helping them defend themselves from antisemitism.

The IHRA definition, proponents would say, is not like a refrigerator or a television, an appliance which can be easily swapped out with little effort or inconvenience if a better model comes along, but more like the plumbing or wiring which exists throughout a house, and is extremely intrusive to replace.  Thus, we should not act as though the definition of antisemitism is forever up in the air, constantly changing, or subject to the whims of small groups of people, but instead operate carefully and in good faith with established definitions as they exist.

IHRA proponents would identify the JDA definition as essentially ephemeral; a thing destined not to function as a durable resource in helping fight antisemitism, but to detract from the existing and future consensus and clarity around the IHRA definition.  They would say that the JDA definition would instead aid in the process of reducing the definition of antisemitism to an endless debate in which little can be done about many instances of Israel-focused antisemitism because too many people adhere to definitions which protect antisemitism from anti-antisemites instead of protecting Jews from antisemitism.

Perhaps the most aggressive critic of the new definition is Lars Fischer, who has specifically argued against a merits analysis of the text of the JDA definition[19], on the grounds that doing so would allow “endless debates about this or that turn of phrase in either definition [to] distract attention from [JDA authors’ and signatories’] outright commitment to the legitimacy of the call for Israel’s destruction.”

The need for a merits analysis of the JDA definition

I do not fully agree with this characterization.  Certainly it is true that we should not lightly change definitions, or fragment our search for a good definition among factions pushing different alternatives.  And certainly there is a high burden for a new definition to surmount, before we should go about urging countries and transnational organizations to disavow or alter the definition they have adopted. 

And in addition, given the long record of racists, sexists, islamophobes and antisemites promulgating minimal definitions of these phenomena in order to protect themselves, the activity of propounding a more-restrictive definition as against a widely adopted and more capacious one is itself somewhat suspect.

And in addition, it feels a little odd for a new definition to be promulgated by surprise.  The JDA definition’s first public airing was as a complete finished document, complete with two hundred signatories and accompanied by simultaneous advocacy in many media outlets around the globe.  No one who wasn’t invited by the organizers had any opportunity to contribute to the process, or even to form an opinion, before it was complete.  In particular, the EUMC and IHRA, which authored and maintain the current definition, do not appear to have even been made aware that a new definition was in work.  And the surprise release happened two days before Passover, when a lot of Jews have other things on their minds.  If they wished to create a document with an “authoritative voice,” which could “unite all forces” against antisemitism, the organizers made a very large mistake by engaging in a secret process and leaving important voices out.

And yet, the document arrived complete with signatures; it must have been made available in advance to hundreds or thousands of people in order to accumulate them.  This speaks of an intentional attempt at secrecy.  Among the signatories, I note Richard Falk, who has had a long series of antisemitism scandals over the last decade plus[20]; David Collier has noted others.  The organizers of the JDA definition thus treated (at least some) notorious antisemites as confidantes, while treating many anti-antisemites as adversaries.  This is perplexing.

The tone of the Declaration is also unnecessarily vituperative.  For example, it accuses the IHRA definition of “undue emphasis” on Israel, Palestine, and Zionism because seven of its eleven examples address them, but then devotes ten of its own fifteen examples, a larger percentage, to the same topics.  It excuses itself for doing so by saying that this was merely a “response” to the emphasis and controversy on this topic in public and scholarly discourse, which is the same reason why the IHRA definition did so!  It’s quite dubious to characterize one definition as “undue” in making this choice, while making the same choice for the same reason.  This does not foster the “frank and respectful discussion” at which the JDA says it is aimed.

It’s also not clear exactly what, other than branding value, makes the Jerusalem Declaration a Jerusalem Declaration particularly.  The Declaration, although advertised as the product of a series of meetings originally convened by the Van Leer Institute in Jerusalem, does not carry its imprimatur or that of the vast bulk of its faculty.  Did they not like the end result?  Of the signatories, only seventeen, of over two hundred, are from Jerusalem.  Of the eight members of the coordinating group, only one is from Jerusalem, and six are from England and Germany.  The process of writing the definition “convened in Jerusalem” in June 2020, but seven of them would have had to fly in, and in the midst of a pandemic they would have had to do a 14-day quarantine which was mandatory at that time.  And another part of the text says the meetings were held online.  Did anyone actually go to Jerusalem?  One part says the process involved “a year of deliberations,” but it began in June 2020, only 9 months before release.  The text was written in English and has been translated into German and Hebrew, but not Arabic, so many of the residents of Jerusalem might have trouble reading the Jerusalem Declaration, or knowing that the copy they have is authentic, given that no one has had the chance to vet the translations.  More clarity about the origin and role of this definition might have been attained with a different name, but then, the “England and Germany declaration on Antisemitism” doesn’t carry quite the same ring, does it?

Nevertheless, I feel that the JDA definition deserves to be addressed thoughtfully and substantively on the basis of its text.  This is for two reasons:

Firstly, as scholars, we owe it to the scholars who created and signed this definition to give it a substantive address.  There are clearly some people of learning and goodwill among them.  They clearly put a lot of work into both preparing this definition and advocating for it, and I shall take their word that they did so in good faith, despite the above.  It must be rewarded with a considered reply, which I shall trust will also be treated with good faith.  If they are wrong, if they have made a mistake, they deserve to hear about it.

Secondly, maybe they are right; we can’t really know until we address their text.  The IHRA definition itself was necessitated almost twenty years ago by a change in the nature of antisemitism to include a greater quantity of Israel-focused antisemitism, and the penetration of this type of antisemitism into more contexts where it was often not recognized.  Perhaps other changes in the landscape since then have made more changes desirable.

We need to substantively evaluate how the JDA and IHRA definitions compare.  Where are they the same?  Where do they differ?  Is the JDA definition actually clearer and more accurate?  Does it better comport with the scholarly understanding of what antisemitism is and how it operates?  Will it make it easier for people and institutions to identify antisemitism in their communities, and avoid tarring legitimate activity?  In the many cases where both definitions agree that a contextual judgment is required, which definition will give the user better guidance about what to look for in making this judgment?  In sum, which definition is actually better for the intended purpose?

If in fact the JDA definition is better in some important way, this doesn’t necessarily mean it should be adopted.  Societies are justified in sticking with conventions even in the presence of better alternatives, when switching is too much bother.  This is why we have imperial units, the QWERTY keyboard, and small bathtubs.  But it would mean that we should consider whether its advantages warrant action when compared to the effort and complexity involved in such action.  For example, such action might include adopting the JDA definition, or, more likely, convening an open, not-secret forum in which to update the IHRA definition.  If the JDA definition is better, we will have to make a decision about what to do.

But on the other hand, if the JDA definition is not actually better, then matters are clear.  We bin it and retain the IHRA definition.

Thus, I will undertake a careful good-faith examination of the comparative text of these two definitions.  I will attempt to determine which is better on purely textual grounds, as if they had been written independently at the same time, and were both new. 

Antisemitism as a species of racism

Before I begin, I will explain my decision not to cover one issue which has been the subject of some comment: the JDA’s strong stance that antisemitism is a species of racism.  Some anti-antisemites, like Lars Fischer, have taken exception to this, because of their belief that antisemitism is not a species of racism but a distinct prejudice which is related to racism. 

I can’t see that the distinction matters much for this purpose.  Fischer says that antisemitism and racism are distinct but related forms of prejudice.  The JDA says that “antisemitism has certain distinctive features” but is a species of racism.  But they also in a sense identify it with every other type of discrimination, saying “the fight against it is inseparable from the overall fight against all forms of racial, ethnic, cultural, religious, and gender discrimination.”  That is, to the extent the JDA views antisemitism as a species of racism, it is only in viewing racism and antisemitism as species of discrimination.  I take this more as a generic statement of support for intersectionalism and allyship than as an analytic claim.

In any event, it doesn’t seem like the distinction matters much to the field as a whole.  Fischer is adamant that antisemitism is not a form of racism, but Hirsh subtitles the Engage blog[21] “the anti-racist campaign against antisemitism.”  There is no consensus on this issue among anti-antisemites of which the JDA could run afoul.

Of course, it is often the case that an unthinking equation of antisemitism to racism has been a tactic of antisemites, who defend themselves against allegations of antisemitism by saying they are not racist and treating the matter as necessarily closed.  But the JDA is not particularly open to this issue, or at least, this is not among its largest flaws.  I will not treat the matter further here.

The positive definitions themselves

Both positive definitions are very brief; the IHRA definition is only 38 words, and the JDA definition is only 16 words.

The IHRA says:  “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.”

The JDA says: “Antisemitism is discrimination, prejudice, hostility or violence against Jews as Jews (or Jewish institutions as Jewish).”

The two definitions are similar in many respects.  However, there are some important differences.

One important difference is that JDA drafters have identified, and even partially corrected, some real awkwardness in the phrasing of the IHRA definition.  The phrasing “a certain perception of Jews” is awkward: which perception?  Surely a definition could clarify the nature of the perception involved somehow.  The JDA addresses this, although poorly, by omitting perception from the equation entirely, and mentioning only actions: discrimination, prejudice, hostility and violence.  In addition, the IHRA definition refers to manifestations directed toward “Jewish or non-Jewish individuals.”  While it clarifies in the examples that non-Jewish individuals can suffer antisemitism if they are perceived as Jewish, this does not relieve the awkwardness of the phrasing in the positive definition.  The JDA addresses this by omitting the possibility of non-Jews suffering antisemitism, which makes the phrasing less awkward, but compromises the definition.  Of course, it is possible envision phrasing which could address these issues without the conceptual changes that the JDA definition makes.

In addition to these phrasing refinements, though, the JDA definition has made some other, less welcome changes to the positive definition.

For one, the JDA definition introduces the awkward formulation “as Jews.”  What is this supposed to mean?  That one could engage in discrimination, prejudice, hostility, or violence against Jews, but be fully blameless of antisemitism because one did not did address them “as Jews?”  How can we tell whether activity is addressed against Jews “as Jews?”  None of the examples address this point at all.  There is language in the IHRA definition which could help address this issue, talking about the selection of targets on account of perceived Jewishness or association with Jews, but the JDA definition contains nothing analogous.

One distressing possibility is that some people will interpret this formulation to mean that antisemitism does not exist in the absence of conscious intent, which is a prevalent mis-definition both in the context of antisemitism and other forms of racism.  This mis-definition omits inadvertent and institutional antisemitism and racism and gives carte blanch to behave in an antisemitic or racist manner as long as one disclaims malicious intent.  This has been an extremely important point in the general social discourse about the definition and operation of racism in recent decades, and a failure to address this point is a major drawback of the JDA definition.

For another example, the IHRA definition is explicit in addressing the fact that manifestations of antisemitism come in both rhetorical and physical forms, and may involve perceptions, not just actions like discrimination, prejudice, hostility and violence.  The JDA definition does not clarify this in the definition, but pushes this important concept to the examples.

Based on a textual analysis, it is clear that despite its awkward phrasings, the extra wording in the IHRA definition is rewarded both with greater comprehension, addressing important forms of antisemitism potentially omitted by the JDA definition, and with greater clarity, making clear that it addresses inadvertent and institutional antisemitism.

The examples: broad similarity

Both definitions offer a very brief positive definition, and rely on a list of examples, most of them focused on Israel, Palestine, and zionism, to clarify matters.  And both definitions rely on a mixture of positive and negative examples.

In some cases, the examples are extremely similar. 

For example, the first example in the JDA definition says “It is racist to essentialize (treat a character trait as inherent) or to make sweeping negative generalizations about a given population,” which is very similar to the second example in the IHRA definition, which speaks of “Making mendacious, dehumanizing, demonizing, or stereotypical allegations about Jews as such or the power of Jews as collective.”

For another, the first example in the IHRA definition is “Calling for, aiding, or justifying the killing or harming of Jews,” while the third JDA example speaks of “assaulting someone because she or he is Jewish, attacking a synagogue, daubing swastikas on Jewish graves, or refusing to hire or promote people because they are Jewish.”

Both definitions contain examples referring to holocaust denial, conspiracy theory, the accusation of dual loyalty, conspiracy theories about finance and geopolitics, collective responsibility, and using the symbols and images of classical antisemitism in criticism of Israel or advocacy for Palestinians.  In general, most of the examples in both definitions map pretty clearly to each other.  This includes all of the first ten examples in the JDA definition, and almost all of the eleven examples in the IHRA definition. 

They differ mainly in the last five examples of the JDA definition (eleven through fifteen).  These examples, unlike those in the IHRA definition, specify what is not antisemitic “on the face of it.”

The IHRA definition also specifies what is not antisemitic, both in its “safe harbor” statement that “criticism of Israel similar to that leveled against any other country cannot be regarded as antisemitic,” and inherently, as definitions, particularly in broad brushstrokes, exclude that which they do not include. 

This clause in the IHRA, along with the positive definition and examples, actually give the ability to construct much of the 11th through 15th examples in the JDA.  For example, the eleventh example in the JDA definition, “Supporting the Palestinian demand for justice and the full grant of their political, national, civil and human rights, as encapsulated in international law,” is not antisemitic under the IHRA definition, both because it fits within the “safe harbor” statement and because it does not run afoul of the definition or examples.  The same is true of most of the 12th and 13th example, and all of the 14th.

In fact, the JDA definition, while it speaks in great hostility of the IHRA’s “blurring” the lines between criticism of Israel and antisemitism in its examples and the “safe harbor” clause, actually not only itself contains almost all the IHRA examples in different phrasings, as discussed above, but also imports the key criterion of the “safe harbor” clause into its definitions repeatedly.  The 11th makes reference to the protection of Palestinian rights in international law (i.e. the same framework that governs other countries).  The 13th specifies that criticism should be “evidence based,” i.e. work under the same epistemological framework as other conflicts, and that “the same norms of debate that apply to other states and to other conflicts over national self-determination apply in the case of Israel and Palestine.”  The 14th says that “Boycott, divestment and sanctions are commonplace, non-violent forms of political protest against states.”  And the 15th discusses the scope of speech protected by UN and European human rights declarations.  To a great extent, the JDA definition, despite its hostility against the “safe harbor” clause, has simply imported it into each of its examples, as its primary justification for them!

That is, in formal, textual terms, there is very little difference between the examples.  The JDA definition displays a curious degree of enthusiasm for declaring things not to be antisemitic, but by a careful interpretation, differs from the IHRA definition little, in all but a few respects.  All the examples above, including the “BDS” clause in the 14th example, actually do fall under the “safe harbor” clause.  There is little in the JDA definition that does not. 

The examples: Distinctions

But what little there is is highly notable, and not in a good way.

First, the JDA definition, in example 12, says that “Criticizing or opposing Zionism as a form of nationalism” is not antisemitic.  This is completely unclear.  Does it mean that it is not antisemitic to criticize Zionism in its capacity as a form of nationalism, together with and in the same terms as one’s criticism of nationalism in general and other forms of nationalism?  If so, then example 12 falls back on the “safe harbor” clause of the IHRA yet again.  But the formulation is unclear, and could be read to permit one to say that Zionism, unlike other forms of nationalism, is bad; to say it is a bad form of nationalism distinct from the others.  It should go without saying that the idea that nationalism is good (or acceptable) for everyone else, and bad (or, unacceptable) for Jews, is antisemitic.  And in fact, of course, “those people, unlike everyone else, ought not to have a nationalism,” is the kind of sweeping negative generalization about Jews which is proscribed in JDA’s example 1.  Or, if you prefer, linking Jews to evil in such a manner that they cannot be trusted with a nationalism (example 2).

This pattern, of declaring something permissible in one example which is proscribed in another, so that the definition as a whole is completely unclear, repeats.

Example 15 states that “political speech does not have to be measured, proportional, tempered, or reasonable” not to be antisemitic, and that criticism which is “excessive or contentious, or…reflecting a “double standard,” is not, in and of itself, antisemitic.”  But example 13 says that criticism should be “evidence based” and that “the same norms of debate that apply to other states and to other conflicts” should apply.  Example 4 speaks of “grossly exaggerating [Israel’s] actual influence.”  Example 2 speaks of conspiracy, and example 1 of sweeping negative statements and says that “what is true of racism in general is true of antisemitism in particular.” 

Is that which is un-measured, disproportional, untempered, unreasonable, excessive, contentious, and reflective of a double standard, not also not evidence based, not in accord with usual debate norms, grossly exaggerative, and sweepingly negative? 

Would we not usually, in the norms that usually control other topics, infer that someone who addresses matters surrounding a particular ethnic group in terms that are un-measured, disproportional, untempered, unreasonable, excessive, contentious, and reflective of a double standard might have some “prejudice” against them (per the positive definition) which we might want to evaluate instead of granting blanket indemnification?  Would that not be true of racism in general?

Is treating people in a manner that is un-measured, disproportional, untempered, unreasonable, excessive, contentious, and reflective of a double standard, not “[hostile]” and “[discriminatory]” (per the positive definition)?

Example 15 declares non-antisemitic a raft of speech and activity which is plainly antisemitic according to both the other examples, and the positive definition, and general principles of anti-racism which the JDA endorses.  This renders the JDA definition completely incoherent.  Anyone who wishes to find something antisemitic can consult the definition and all the other examples, and anyone who wishes to find it not-antisemitic can wear example 15 (in combination with 11-14) as a shield.  Someone who wishes to carefully interpret the whole text will find nothing but befuddlement; the text is befuddled.

Because of this very befuddlement, the authors seem very certain their text will be misinterpreted.  The 14th example, which has indeed been controversial, states simply that “Boycott, divestment and sanctions are commonplace, non-violent forms of political protest against states. In the Israeli case they are not, in and of themselves, antisemitic.”  This allows implicitly that some examples of this behavior are antisemitic, and gives a criterion similar to the “safe harbor” clause in the IHRA to determine when.  But the authors also give an FAQ with two clarifying questions on the 14th example.  One asks “But doesn’t guideline 14 support BDS as a strategy or tactic aimed against Israel?”  The answer says “Guideline 14 says only that boycotts, divestments and sanctions aimed at Israel, however contentious, are not, in and of themselves, antisemitic.”  But even this is not enough; they need a second question!  This asks “So, how can someone know when BDS (or any other measure) is antisemitic?”  The answer to the FAQ question refers the reader to the examples and asks them to make a judgment call.

Of course, if it’s OK to ask people to look at the text and make a judgment call, then example 14 is unnecessary.  The “safe harbor” clause of the IHRA, or the similar formulations seen in numerous parts of the JDA, ultimately amount to the same standard.  But the authors of the JDA wanted to make a specific and permissive sounding mention of boycott, divestment and sanctions so much that they simply had to do it, even though it was unnecessary, and even though they themselves knew it would cause so much confusion as to necessitate two FAQ questions.

Similarly, the status of antizionism is also the subject of two FAQ questions, because the authors simply could not bear to make their definition say clearly that antizionism is often antisemitic and make a clear statement of the standards by which to judge when it is.  Example 10 says that it is antisemitic to deny “the right of Jews in the State of Israel to exist and flourish, collectively and individually, as Jews, in accordance with the principle of equality.”  Would that include national flourishing?  The IHRA would say so; it speaks of the Jewish people having a “right to self-determination” in example 7.  And the JDA seems to think so as regards Palestinians; example 11 refers to Palestinians’ “political, national, civil and human rights.”  Do Jews also have political and national rights?  According to the JDA, apparently not; any arrangement which offers “full equality to all inhabitants” is considered acceptable in example 12.  But what about the right to “flourish collectively” in example 10?  If Palestinians get national rights, and Jews only get the right to exist and flourish, is that a double standard?  Don’t worry; double standards are fine (example 15).

The JDA is unclear and inconsistent, not incidentally, but fundamentally, because of these changes.  And while it is difficult to peer into the motives which lie behind a document, it’s difficult to discern why it would have been written this way, if not to excuse the antisemitic behavior it excuses, which would require the document to be unclear and inconsistent.  For whatever reason, the authors could not bear to make their meaning clear, e.g. not as clear as the IHRA definition.

The real world impact of the JDA definition

And this lack of clarity has been fully and immediately expressed in the real world responses to the JDA.  The primary strain of press and institutional coverage of the JDA definition has been centered around claiming that the JDA definition says the BDS movement is not antisemitic.  But the JDA doesn’t do that!  They went out of their way to sound like they were doing that with the phrasing of example 14, but to clarify in the FAQs that they were not.  This confusing way of writing has been rewarded by confusion.

In the Palestine Chronicle[22], with the headline “‘Jerusalem Declaration’ Insists BDS is Not Anti-Semitic, in Disagreement with IHRA Definition.”

In the Jewish Journal[23], with the headline “Jewish Scholars Unveil New Anti-Semitism Definition Saying BDS Isn’t Anti-Semitic.” 

In the Forward[24], with the headline “Leading Jewish scholars say BDS, one-state solution are not antisemitic.”

In the Jerusalem Post[25], with the headline “Over 200 scholars create new antisemitism definition excluding BDS.”

In the Electronic Intifada[26], with the statement that “The new declaration states that…support for the boycott, divestment and sanctions (BDS) campaign [is] not anti-Semitic.”

In Mondoweiss[27], with the statement that “Guidelines 10-15…are a clear statement that support for BDS has nothing to do with antisemitism.”

By the Jewish Voice for Labour[28], with the statement that “the JDA does provide sanction for…support for the nonviolent BDS movement and its tactics.”

These examples span the ideological spectrum, and come from the USA, Europe, Israel, and Palestine.  And they all make the same mistake that the JDA group sought to guard against with not just one, but two FAQs appended to its main document.  That is, this confusion is not just something they should have anticipated, but something they did anticipate.  And yet it was the primary reaction of the press to the new definition!

For a definition whose raison d’etre is supposedly greater clarity, this is a disaster.  It is much like an airliner, designed for greater safety, that crashes and kills its passengers on the maiden flight.

Joshua Shanes and Dov Waxman, two JDA signatories, make this confusion themselves in their Slate op-ed supporting the launch of the JDA definition[29].  In it, they say “No doubt, the JDA, like the IHRA definition, will elicit controversy and criticism, especially from those who insist that all anti-Zionism is inherently antisemitic or that the BDS movement is incorrigibly anti-Semitic.”  But of course, neither the IHRA nor JDA definitions say that all antizionism is antisemitic, or that boycott, divestment, and sanctions as tactics are, nor does the JDA definition say that the BDS movement is not.

Barry Trachtenberg, another JDA signatory, did the same in Jewish Currents[30].  He says that the JDA definition is “declaring that ‘criticizing or opposing Zionism as a form of nationalism’ is prima facie not antisemitic, nor is the BDS movement.”

Alon Confino, another signatory, did the same in a video released on the JDA website accompanying the definition[31].  “Our signatories,” he said, “have a range of views about the BDS.  The entire gamut of views.  But they share one principle: that the BDS, as a non-violent action by civil society opposing occupation in lines with other boycott movements in the 20th century is not antisemitic.”

Even Brian Klug, one of the members of the JDA coordinating group, wrote an article in The Nation supporting the JDA launch[32], in which he says that “people invoke the IHRA definition to claim that applying the term “apartheid” to Israel and supporting BDS are intrinsically anti-Semitic. But they are not.”  Klug himself says that “supporting BDS” is not inherently antisemitic.  Not that boycott, divestment, and sanctions as tactics are not inherently antisemitic, but that BDS the movement is actually not antisemitic. 

This is the distinction to which the JDA devotes two clarifying FAQs, but multiple JDA signatories and a member of the JDA coordinating group cannot keep it clear themselves while writing published articles supporting the launch of the JDA definition!  What hope could the rest of us have?

Worst of all is the response of the BDS movement.[33]  The website of the movement’s central committee posted a long comment only hours after the JDA document was published (did they know it was coming?).  They call the JDA definition “coherent and accurate,” hail it for treating antisemitism as a species of racism, and identify lots of things they like about its examples and where they differ from the IHRA (although most of them, as I point out above, don’t).  Then it says that “the JDA recognizes as legitimate free speech…support for the nonviolent BDS movement and its tactics.” 

The BDS statement then identifies “flaws” in the JDA definition, including this: “the JDA’s ‘guidelines’ still try to police some speech critical of Israel’s policies and practices.”  That is, they are upset that any type of antizionism is identified as antisemitic.  The BDS statement gives three examples of things it wishes not to be considered antisemitic which are declared antisemitic by the JDA definition: “Portraying Israel as the ultimate evil or grossly exaggerating its actual influence,” (from example 4) “Applying the symbols, images and negative stereotypes of classical antisemitism … to the State of Israel,” (example 6), and “Denying the right of Jews in the State of Israel to exist and flourish, collectively and individually, as Jews, in accordance with the principle of equality” (example 10).

Think what the BDS movement is saying here.  We wish, they are saying, to portray Israel as the ultimate evil, and to grossly exaggerate its actual influence, by applying the symbols, images, and stereotypes of classical antisemitism to Israel, in order to deny the collective right of Jews to exist and flourish collectively.  We know full well that the JDA definition says these things are antisemitic.  And we support the JDA definition of antisemitism because, despite recognizing these things as antisemitic, the new definition will facilitate our attempt to do these things without their being recognized as antisemitic.

The real world impact of the JDA definition has been revealed very clearly.  The textual incoherence of the definition has been rewarded with total confusion in every corner.  No one, no one, no one, has taken the JDA in the manner for which its text calls; not even the drafters.  The primary public reaction has been the exact kind of confusion which the IHRA definition allegedly fosters, and which the JDA definition allegedly clears up. 

It is not a defense against this issue to claim that the JDA definition is clear and has been misinterpreted.  This is for three reasons:

  1. The argument against the IHRA, and for the necessity of the JDA definition, is that although the IHRA definition largely operates correctly, it is open to misinterpretation.  The JDA says that the IHRA is “unclear in key respects and widely open to different interpretations…Whatever the intentions of its proponents, it blurs the difference between antisemitic speech and legitimate criticism of Israel and Zionism.”  Shanes and Waxman say “Although it does not simply equate anti-Zionism with anti-Semitism, or label all criticism of Israel to be anti-Semitic—as some opponents of the definition assert—its vague, conditional wording is open to misinterpretations and misuse. Its conditional phrasing…is too often forgotten, or even purposefully ignored.”  Klug says “Whatever the IHRA authors intended, this polarization is a fatal indictment of their definition.”  That is, the JDA declares that a definition being open to misinterpretation is a fatal blow.
  • In fact the JDA definition is textually unclear, in addition to having been misinterpreted.
  • Even the drafters and signatories of the JDA have been party to the confusion in the press around its launch.  This confusion has not just affected some people, but functionally everyone.  If no one at all can correctly interpret the text, then the issue lies with the text, not the interpreters.

As an attempt at clarity, the JDA is necessarily for textual reasons, and has been in actuality upon its launch, a complete, irredeemable failure.


The JDA definition of antisemitism is a failure.  The primary effect of the definition’s launch has been to create extra confusion and controversy about antisemitism, to the visible delight of antisemites and the dismay of anti-antisemites.  This issue is not just about misinterpretation; my close analysis has shown that it is inherent in the text of the JDA definition.  Despite being, in formal terms, very similar to its bete noir the IHRA, the differences between them both make misinterpretation more likely, and introduce fundamental conceptual incoherence.  It is a very bad definition, much worse than the IHRA definition, and it is hurting the struggle against antisemitism.

It is a pity that these issues could not have been pointed out before, because of the secrecy of the drafting process.  The drafters chose to treat some antisemites as confidantes, many anti-antisemites as adversaries, and Jerusalem as a branding accessory.  That antisemites would be thrilled, anti-antisemites dismayed, and Jerusalem confused, has been the unavoidable result.

Accordingly, the signatories, witnessing the critical analysis of the text that was unavailable during drafting, and the universal confusion on release which confirms that analysis, should consider the issue carefully and then recant their endorsement of the definition.

Institutions should not adopt it.  They should continue to use the IHRA definition.

People should not use it.  They should use the IHRA definition, the dictionary, or common sense.

The JDA definition of antisemitism should be treated by everyone as dead on arrival.

It’s lamentable that this whole affair ever happened.  Jews should never have been subjected to the spectacle of watching antisemites gloat about a group of scholars and their new definition of antisemitism facilitating the project to weaponize the symbols of classical antisemitism against modern-day Jews by way of Israel.

This is not to say that the IHRA definition is perfect or that the work of defining antisemitism is over.  The IHRA definition itself was a reaction to a change in the types of antisemitism that existed in the world at the time it was written, and it is certain that antisemitism has changed further since then and will continue changing.  There is real and important awkwardness in the IHRA’s positive definition, for example, which the JDA definition at least tries to correct.  We may need new definitions of antisemitism in the future; perhaps we need them now.  And the process of applying existing definitions to events in the world is itself an act of definition, an ongoing process, a humane and political one.

But the IHRA definition’s main insight, that anti-Israel antisemitism was becoming increasingly common and increasingly significant as a portion of global antisemitism, was true when it was written and has become truer now.  The importance of being able to recognize it has only increased.  And the ability to do that, and to avoid tarring legitimate criticism of Israel and support for Palestinians, is much stronger with the IHRA definition (including examples) than the JDA definition and its examples, both because of its long history and wide institutional and academic support, and because the text of the IHRA definition is better than the JDA text.

If, some day, or if, now, we need a new definition of antisemitism, anyone seeking to write it should avoid the mistakes made by the JDA drafters.  They should seek to build on existing definitions, not vituperate against them and exaggerate their flaws.  They should make an open process, not a secret one.  They should treat anti-antisemites as allies and antisemites as rivals, not the reverse.  And they should be careful to produce a clear and consistent text, not a disastrously incoherent one.  There are a lot of challenges in formulating definitions for slippery concepts like this, and not every attempt to write one will be very successful at navigating those challenges.  But every attempt at writing a definition should be able to avoid basic procedural mistakes like these, and should be able to avoid releasing such a fatally flawed finished product.

Ari Allyn-Feuer


































Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s