In the wake of the EHRC report on antisemitism in the UK Labour Party, a variety of complaints have been made about the IHRA definition of antisemitism. Two of the critics’ principal claims are first, that the definition doesn’t tell us precisely and specifically which people, language, acts and practices are antisemitic; and second, that it suppresses freedom of speech. Philosopher Eve Garrard explains why the first of these objections is true but irrelevant, while the second isn’t even true.
Defining antisemitism isn’t easy. Its shape-shifting nature ensures that it’s hard to find a definition sufficiently complex to fit its long and blood-stained history. So it’s quite true that the IHRA definition doesn’t provide us with a philosophically satisfactory account of antisemitism. The philosophical hunt for the necessary and sufficient conditions for an object or phenomenon to fall under the term being defined is generally a very lengthy business. It’s also quite often a failure. (The term ‘knowledge’, so central to the practice of philosophy, is perhaps the most striking, and of course highly-contested, example.)
Back in the last century, the philosopher Ludwig Wittgenstein took serious issue with this approach to definitions. He pointed out that for some concepts, there simply are no necessary and sufficient conditions for their use – that is, there are no essential features whose presence ensures that the concept applies, and whose absence ensures that it doesn’t. His example was the idea of a game, a term which all speakers of the English language can use correctly and effortlessly. But there are no necessary and sufficient conditions for its correct use: the idea of a game covers a very wide range of activities, but there is no common essence which all games possess. And quite a lot of our concepts are like this. There is no feature which all cases of antisemitism possess, and which is not possessed by anything which isn’t antisemitic. But there are many features which are possessed by many cases of antisemitism, and many other features which are possessed by other cases, and there is much overlap between them. We can cite some of the central cases, and use them to illuminate the more peripheral ones.[i] For some concepts, that’s the best we can do by way of definition. And this is how the IHRA definition treats antisemitism.
Is this kind of definition useful? Yes, very useful, so long as we’re not looking for a tight philosophical definition. It’s useful because it helps people to see what kind of thing antisemitism is, and thereby inform their judgement on new cases which may come their way, and adjust their behaviour accordingly. That is, it’s politically useful; it helps us to understand past examples, and to adjudicate new conflicts. (It’s not only antisemitism which raises this issue: racism and sexism are other concepts where we are unlikely to find an unchanging essence present in every case.) Does this mean that the IHRA definition of antisemitism won’t do all our judging for us? Yes, it does mean that; we’ll still have to work out which cases of criticism of Israel, for example, actually amount to antisemitism. The IHRA definition, particularly in the examples it provides, alerts us to the fact that antisemitism is in the offing; but our own moral capacities, and sensitivity to the individual context, will still be needed to tell us what we should say or do in the particular context we’re facing. But that’s what morality is like: simple straightforward moral rules can only take us so far; to work out how they apply to the case in front of us, we have to think for ourselves. (That’s why, for example, the moral rule that tells us that it’s wrong to kill innocent people won’t by itself settle the question of whether euthanasia can ever be justified.)
On Freedom of Speech
So the complaint about the IHRA definition’s failure to provide an unchanging essence for all cases of antisemitism, though true, is irrelevant – too many of our other concepts are in the same boat for this complaint to count against IHRA. However, the other (putatively moral) objection – the claim that IHRA underpins an attack on the freedom of speech of critics of Israel – is not even true. Here’s why: the IHRA definition is peppered with conditional verbs, saying that this or that ‘may’ be antisemitic, or ‘could’ or ‘might’ be antisemitic. Its list of examples is prefaced by the remark that they ‘could’ provide cases of contemporary antisemitism, but that such antisemitism is ‘not limited’ to just those examples. But to say that a practice may be antisemitic is to allow that it may not be. To say that applying double standards to Israel could in some cases be antisemitic leaves room for the possibility that in some cases it isn’t antisemitic. That’s how these conditional verbs work. And the reason that we need words that work in this cautious way is that racism of any kind occurs in the complicated contexts of our moral lives, and good moral judgements are highly context-sensitive. So the charge that the IHRA definition threatens our freedom of speech simply isn’t true; what the definition does do is alert us to the fact that some ways of talking about Israel are antisemitic. The only view which this definition threatens is the view that criticism of Israel can never, ever, in any circumstances, be antisemitic. But this is not a view which is even remotely plausible (although some critics of the IHRA definition do seem to find it attractive). It is, of course, always possible that the IHRA text could be misused to assert the mistaken claim that criticism of Israel is always antisemitic. Misuse is a possibility with any text, but here the IHRA definition itself, with its cautious conditional claims, protects us all from accepting either of these implausible views.[ii]
None of these thoughts are new ones; so why are some people (such as David Feldman, writing in the Guardian)[iii] so hostile to the IHRA definition? Here’s one possible explanation: people who are, perhaps unconsciously, in the grip of the idea that there must be a single essential core to cases of antisemitism may be committed to an account of antisemitism as being hatred of Jews as Jews. This is a very unsatisfactory definition, as can be seen if you try it out on cases of other varieties of racism or indeed sexism. Back in the day, hatred would have been far too strong an emotion to impute to the kind of person who didn’t want Jews in the golf club; it was far more likely that he felt an uneasy distaste for people who weren’t quite like himself. But blackballing Jews from the club was certainly a case of antisemitism, though hardly of a very important kind. Hatred of Jews isn’t necessary for antisemitism – it can result from inaccurate and demeaning stereotypes, or even from the appeal to traditional customs or speech.
Another candidate for the position of the (supposedly) essential characteristic of antisemitism, without which it just wouldn’t be antisemitism at all, is thought by some to be the presence of antisemitic intent. Leave aside the problem that if we can’t find an independent definition of antisemitism then we’re unlikely to be able to find a definition of antisemitic intent. There’s an even more pressing problem for the definition in terms of intent: it’s the fact that there are cases of antisemitism which don’t involve discriminatory intention at all.
On Institutional Antisemitism
A striking example of the presence of antisemitism in the absence of either hatred against Jews as Jews, or antisemitic intent, can be found in cases of institutional antisemitism.[iv] When an organisation – a business, say, or a public service or a political party – has practices or policies which significantly disadvantage Jews for no good reason (very important clause) then its behaviour is institutionally antisemitic.
Why do we need the special concept of institutional antisemitism to mark off this form of antisemitism from the more familiar ones? Why not just call them all antisemitism? One reason is that usually antisemitism is a matter of individuals, singly or in groups, deliberately discriminating against Jews. But in cases of institutional antisemitism no individual member of the institution need have deliberately and knowingly singled out Jews for unfair treatment: it’s the way the overall institution operates which creates the discriminatory impact. So, for example, if an organisation decides to hold all of its most important policy and career promotion meetings on Saturday mornings (when no observant Jews would be able to attend), and there’s no good reason for this timing of the meetings, then this looks like a case of institutional antisemitism.
There’s a second notable feature of institutional antisemitism: no hatred, indeed no hostile feelings at all, need be involved. (They may be involved, they often are involved, but they needn’t be.) The more ordinary cases of antisemitism do very frequently stem from strongly hostile feelings, so people often think that where there’s no hatred, there can be no antisemitism. But this really doesn’t follow: if there’s institutional behaviour which discriminates against Jews, which disadvantages them compared to other people without any good reason, then this is antisemitic behaviour. It’s a form of antisemitism which doesn’t depend on feelings at all, which is why people who are accused of taking part in institutional antisemitism often flatly refuse to believe it, on the grounds that they have no hostile feelings towards Jews. But it’s the discriminatory effect, not the presence or absence of hostile feelings, that makes the behaviour or policy institutionally antisemitic.
This is a quite general point about discrimination, and it applies to discrimination against persons of colour as well as against Jews, and for that matter it also applies to discrimination against women and to discrimination on the basis of class. If a school were to have a policy of discouraging black pupils, or girls, or children from deprived backgrounds, from applying to top universities, on the (misplaced) grounds that they probably wouldn’t be accepted, and in any case would be more comfortable in a less academically demanding environment, then this could seriously disadvantage such pupils. But the teachers who would have devised and implemented the policy are unlikely to have been motivated by hatred at all, but more plausibly by a combination of an indulgent but misplaced desire to protect their pupils from rejection, combined with recourse to unexamined and misleading stereotypes. (Such things have happened in the past, and may still be happening in some places now.) This is how policies can be adopted which unfairly disadvantage certain groups of people, without there being any hatred or hostility at all in play.
The ‘without good reason’ clause really matters here. This is because there are some cases where members of one racial group do get treated more advantageously than members of another group, but if there’s a good reason for the differential treatment, such as an urgent need for a distinctive medical treatment disproportionately needed by members of that group, then no racism, institutional or otherwise, need be involved. However in the absence of a good reason, policies or practices which disadvantage Jews compared to people of other ethnicities will be cases of institutional antisemitism.
One of the most troubled areas of dispute about antisemitism concerns attitudes and behaviour towards Israel, the world’s only Jewish state, whose existence is supported by a very large majority of the world’s Jews. It’s often claimed, quite truthfully, that criticism of Israel and its supporters isn’t necessarily antisemitic; as we have seen, this is in accord with the IHRA definition. Some criticism of Israel isn’t antisemitic, and that’s a fact. It’s also a fact that other cases of criticism of Israel really are genuinely antisemitic. So what makes the difference? One principal difference emerges when criticism of Israel, and the policies and practices which sometimes go along with it such as a boycott of Israeli products and institutions, unfairly disadvantage Jews in comparison to other groups. So we need to look at how other groups get treated. Do people call for boycotts against, or campaign for the elimination of, other countries which have comparable, and often far worse, human rights records than Israel does? This very, very rarely happens. If there’s no good reason for singling out Israel for especially adverse treatment, up to and including calls for its annihilation, then such treatment by organisations would be a striking example of institutional antisemitism, particularly because it’s far more likely to have an adverse impact on Jews in this country than on Israel itself.
The IHRA definition draws our attention to the kinds of speech and action which are liable to be antisemitic. It also draws our attention to the complexity of such cases, to their sensitivity to context. It discourages us from the kind of simple-minded moral-purity-seeking approach which says that certain types of action are either always antisemitic, or they’re never antisemitic. In this the IHRA definition is correct, and it stands as a reminder of how complicated our behaviour, and our lives, are.
[i] See, for example, John Finnis, Natural Law and Natural Rights (Oxford University Press, 1980, 2nd edn 2011) for a development of this kind of view.
[ii] For further rigorous discussion of these issues, see Bernard Harrison and Lesley Klaff, ‘The IHRA Definition and Its Critics,’ in Contending with Antisemitism in a Rapidly Changing Political Climate, ed. Alvin H. Rosenfeld, (Indiana University Press 2021) (forthcoming.) Also see their ‘In Defence of the IHRA Definition’ in Fathom, January 2020.
[iii] The Guardian, 2nd December 2020
[iv] The following material draws on a fuller discussion of institutional antisemitism which appeared in Forced Out: Labour antisemitism resignation letters, ed. Judith Ornstein, Kitty Hawk Press, 2019].