Why context matters: Constitutional Court leaves the hate speech threshold open to judges’ interpretation

Last week, the Constitutional Court finally handed down judgment in a hate speech case argued before that court almost 2½ years ago. But for those of us who had hoped that the court’s earlier hate speech judgment in the Jon Qwelane matter would provide guidance on what constitutes hate speech and would lead to more predictable outcomes in hate speech cases, the recent judgment comes as a bit of a disappointment.

It was somewhat surprising that a unanimous Constitutional Court last week endorsed a finding by an Equality Court that a vicious attack on Zionism and Zionists constituted hate speech based on Jewish ethnicity and religion because the offending speaker mentioned Hitler in his attack on Zionists. I had assumed that the Constitutional Court would accept the view (also advanced by experts testifying before the Equality Court) that criticism of Zionism, the state of Israel, or its government cannot, per se, be equated with anti-Semitism. (I assume many Zionists and supporters of Zionism and the Israeli state and/or government might see things differently.)

In South Africa, few legal-political questions elicit as much heat and as little light as arguments about whether specific controversial speech acts constitute hate speech or not. This is not only because extreme political polarisation and widespread intolerance of differences in our society stir up irrational passions on such matters. It is also because the concept is not well understood by most people who participate in these “arguments”. (Many people wrongly conflate “hate speech” with speech that they find disagreeable, hurtful or irritating, or with utterances made by politicians or other public figures they disagree with or actively hate.)

It has not helped that different courts had often interpreted and applied the hate speech provision in section 10 of the Promotion of Equality and Prevention of Unfair Discrimination Act (Pepuda) in radically different ways, leading to radically different outcomes in cases dealing with similar kinds of speech. It is perhaps telling that the hate speech judgment delivered during this period that really got to the heart of the matter was written not by a judge of any of our higher courts, but by DM Thulare, the Chief Magistrate of Cape Town. In that judgment, the court rejected the argument that a work of art titled Fuck White People was hate speech, at least in part because the magistrate concluded that a reasonable person would have understood that the work of art was intended to spark self-reflection and positive change, not to cause harm.

The irrational fury caused by the Fuck White People artwork and many other forms of challenging expression also suggests that when people argue about what constitutes “hate speech” and what does not, they often seem to be arguing about (or also about) something else entirely: about history and how the past should be remembered or forgotten; about dignity, identity and belonging; about power and the loss of power; about responsibility and the avoidance of responsibility; and, of course, about whose grievances, prejudices, hatreds and fears should weigh heavier in the eyes of the law.

I had assumed that the Constitutional Court’s judgment in Qwelane v South African Human Rights Commission (handed down late last year), in which that court provided a “definitive” interpretation of section 10 of Pepuda would clarify the legal position on hate speech, and that this would reduce the confusion – at least among lawyers – about what kinds of speech could reasonably be expected to amount to hate speech. But the most recent judgment of the Constitutional Court in SAHRC obo South African Jewish Board of Deputies v Masuku suggests that the outcome of hate speech cases will remain difficult to predict. Not that this will deter, say, outraged AfriForum leaders and follows from pretending that it is outrageous for anyone to suggest that their most recent “Kill the Boer” case against Julius Malema might not be the slam dunk win they think (or pretend to think) it is.

After Qwelane, section 10(1) of Pepuda prohibits any person from publishing, propagating advocating or communicating “words that are based on one or more of the prohibited grounds, against any person, that could reasonably be construed to demonstrate a clear intention to be harmful or to incite harm and to promote or propagate hatred”.

Section 10 only regulates speech that targets a group based on specific listed and analogous grounds such as race, sex, gender, sexual orientation, religion ethnicity, language, culture, HIV status and the like. It does not regulate speech that targets “unprotected” groups such as farmers, accountants, politicians or, for that matter, AfriForum or EFF members. Moreover, the test is objective, which means how the members of the targeted group may have viewed or experienced the speech is irrelevant. The test is also fact and situation specific. The same words (or song) could amount to hate speech in one context, and not amount to hate speech in another. This is why it would be wrong for a court to impose a blanket ban on the song or other types of speech.

As the Constitutional Court confirmed in its Masuku judgment, a reasonable person would consider “who the speaker is, the context in which the speech occurred and its impact, as well as the likelihood of inflicting harm and propagating hatred” against a protected group, as well as the broader historical context, including “the reality of our past of institutionally entrenched racism”. This is why, depending on the context and other circumstances, it is more likely that a person wearing a “Fuck Black People” T-shirt would be guilty of hate speech than a person wearing a “Fuck White People” T-shirt. (But where the T-shirt is worn, who it is worn by, and other relevant factors pointing to the intention of the wearer will obviously also be important.)

Hate speech cases are particularly difficult to resolve (and the outcome difficult to predict) when the speech does not explicitly target a protected group. This problem arose in the Masuku case as Masuku did not explicitly target people because of their Jewish religion or ethnicity (protected groups), but instead targeted Zionists. In Masuku the Constitutional Court accepted that a reasonable person would not conclude that people were targeted because of their Jewish religion or ethnicity, merely because the majority of Jewish people were also Zionists. Something more is needed. What made the difference in this case was the reference to “Hitler” which, the court held, would have led a reasonable person to conclude that the statement was based on Jewish ethnicity and not merely on a Zionist political orientation.

It is interesting to note that the court held that a reasonable person would not have concluded that other statements (including one referring to any family “who sends its son or daughter to be part of the Israel Defence Force (IDF)”) had targeted individuals based on their Jewish religion or ethnicity, despite the fact that “only Jewish families would send their children to join the IDF”. The court justified this conclusion by pointing out that it was unlikely that a Jewish person would join the IDF if they were not a Zionist supporter.

One way to make sense of this conclusion is to assume that statements targeting a non-protected group (including Zionists) would not constitute hate speech merely because the majority of that group are also members of a protected group (including Jewish people). Other evidence would be required for a reasonable person to conclude that an attack on Zionists was in fact an attack on Jewish people. But it is not that clear from either the Qwelane judgment or the Masuku judgment exactly what kind of evidence would suffice.

A party like AfriForum that complains about the singing of a song targeting farmers, or perhaps apartheid supporters, or perhaps white supremacists, will have to convince the court that there are other factors that would convince a reasonable person that the singing of that song at a particular event in fact targeted a racial group and not a group of people based on their political views. It would then further have to convince the court that given the specific event where the song was sung, the identity of the person who sung it, and other relevant factors, a reasonable person would conclude that the intention of the singer was to be harmful or to incite harm and to promote or propagate hatred against white people.

I don’t know how the court will rule in this instance, not only because I have not watched all the court proceedings, but also because the Constitutional Court test leaves quite a bit of discretion to an individual judge when applying the test. But what I do know is that anyone who claims to be absolutely certain about the outcome of this, or any other hate speech matter, probably has not read the relevant Constitutional Court judgments, or does not understand (or pretends not to understand) what the test for hate speech actually is.

 

BACK TO TOP