Follow AfricanCrisis on Telegram
You can follow AfricanCrisis‘s posts directly on Telegram. All the posts and videos go out on this Telegram Channel.
[Here you can see a legal analysis. Is this valid? Or is it simply a case of the Blacks doing as they please? Jan]
In HOA for AfriForum, SC says songs at issue clearly constitute hate speech in terms of section 10(1) of the Equality Act
IN THE SUPREME COURT OF APPEAL (BLOEMFONTEIN)
SCA case number: HC case number: EQ 04/2020
In the appeal between:
AFRIFORUM – Appellant
(applicant a quo)
ECONOMIC FREEDOM FIGHTERS – First respondent (same a quo)
JULIUS SELLO MALEMA – Second respondent (same a quo)
MBUYISENI NDLOZI – Third respondent (same a quo)
APPELLANT’S HEADS OF ARGUMENT
1. This appeal concerns whether the singing of the songs, Dubula ibhunu (Kill/Kiss the Boer) and Bizani iFire Brigade (with its central lyric, Burn the Boer), by the second and third respondents in the circumstances which are common cause in this matter constitute hate speech in terms of section 10(1) of the Equality Act, 4 of 2000 (‘the Equality Act’).
2. Applying the section 10 test – as reformulated provisionally, pending Parliament’s amendment1 – it is submitted that they unequivocally do. The elements of hate speech are all present. Both songs communicate words that are based on one or more of the prohibited grounds (race, social or ethnic origin) against persons, in this case ‘Boere’, that could reasonably be construed to demonstrate a clear intention to be harmful or to incite harm (to shoot, kill or burn them) and to promote or propagate hatred.
3. The High Court, sitting as the Equality Court (per Molahlehi J) found that the songs do not constitute hate speech and dismissed the application.2 This appeal proceeds with its leave.
4. The issue in this appeal is whether this finding is legally correct. The facts – the singing of the songs in circumstances elaborated below, and not in material dispute – are not in dispute. Hence the irrelevance of the overwhelming bulk of the record. (The appellant had proposed its substantial elimination; the respondents refused. The attached exchange of letters between the attorneys confirms this). We revert to this aspect in dealing with costs.
THE FACTS – THE SINGING OF THE SONGS
The first song: ‘Kill/Kiss the Boer’
5. The first impugned song or chant (containing the interchangeable – see para  of the judgment – lyric, ‘Kill/Kiss the Boer’)3, was sung by the second respondent, Mr Malema, the leader of the EFF on at least six occasions.4
6. On 7 November 2016 during an EFF rally in Newcastle Mr Malema sang or chanted the song ‘Dubul’ ibhunu’ hereafter ‘Kill the Boer’. From video footage he can be heard singing ‘shoot to kill’, ‘kiss the Boer’ and ‘kiss the farmer’ – this while visibly gesturing with his hand the pointing and shooting of a firearm.
7. On 30 July 2017 on the EFF’s fourth birthday celebration in Durban Mr Malema led the singing of ‘Kill the Boer’.
8. On 25 May 2018 the EFF organised the Africa Day Celebrations Event where Mr Malema led the singing of ‘Kill the Boer’.
9. On 29 October 2018 at the Vaal University of Technology Mr Malema led the singing of ‘Kill the Boer’ after addressing hundreds of students. He can once again be seen gesturing the pointing and shooting of a firearm with his hands.
10. On 2 February 2019 at the EFF manifesto launch in Shoshanguve Mr Malema led and encouraged the singing of ‘Kill the Boer’.
11. On 15 April 2019 at kwaTsheka Sports Ground, eNseleni Mr Malema led and encouraged the singing of ‘Kill the Boer’ at an EFF event.
12. The court a quo explicitly found that the first song comprises communicated words, within the purview of section 10.5 Implicitly it held the same as regards the second impugned song (containing the lyric ‘Burn these Boers’). It further found as an undisputed fact that during his chanting of the former Mr. Malema used ‘the gesture of a gun in his hands’ (a matter not mentioned further in the judgment, and to the importance of which we revert).6 To the words would properly have to be added, on the authorities cited below, in relation to the first song, the accompanying gestures: the words were used, and would be understood, in a physical context.
The second song: ‘Burn the Boer’
13. In Senekal in early October 2020, a 21-year-old farm-manager Brendin Horner was brutally killed and his corpse strung up on a fence. Two suspects were arrested and on 6 October 2020 distraught members of the local community, agricultural and political leaders and civil society organisations protested against the murder.
14. On 16 October 2020, in tense circumstances prevailing in Senekal members of the first respondent (the ‘EFF’) arrived and chanted ‘Kill the Boer, kill the farmer’ to a community grieving the killing.7
15. It was here, in these circumstances, that the third respondent’s then national spokesperson, the second respondent ( Dr Ndlozi) sang the following song:
Bizani Fire iBrigade Bath’umlilo lo Bizani iFire Brigade Shisa lamabhunu
EFF Ingen’endaweni Translation:
Call, the Fire Brigade This is fire
Call, the Fire Brigade
Burn these Boers
EFF enters the space/place8
THE LAW – THE SECTION 10 REQUIREMENTS FOR HATE SPEECH
16. In Qwelane9 the Constitutional Court determined that – pending amendment by Parliament – section 10(1) of the Equality Act should read, in relevant part, as follows:
‘Subject to the proviso in section 12, no person may…communicate 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.’10
17. The ‘prohibited grounds’ are those defined as such in section 1 of the Equality Act (which include – relevant to the present matter – race, ethnic or social origin, or any other ground undermining human dignity or adversely affecting rights in a manner comparable to the main listed grounds).
18. It is submitted that the post-Qwelane section 10 is based on two objective inquiries. The first inquiry (once the Court found that words were communicated,
within the purview of the provision) is whether those words were ‘based on’ what are defined in section 1 of the Equality Act as ‘prohibited grounds.’ This is an objective test. It is for a court to decide whether the words used are ‘based on’ the one or more of the 17 social stigmata listed in sub-section (a), or the yet- widening provisions of (b). The answer to that inquiry is not determined by what
the speaker intended (least of all by what the speaker might volubly attest they intended). The speaker’s intent becomes relevant – and this only on the basis of what is reasonably to be construed from the words as to the intent – in the
second part of the (Qwelane-reformulated) section 10 (quoted in para 16 above).
19. The second inquiry is whether the words used ‘could reasonably be construed to demonstrate a clear intention to be harmful or to incite harm and to promote or propagate hatred.’ This test too is objective. The subjective intent of the author is irrelevant. As the Constitutional Court confirmed in paragraph  of Qwelane: ‘This approach accords with the interpretation advanced in SAHRC v Khumalo that “[t]he objective test in section 10(1) implies in the terminology used to articulate it, that an intention shall be deemed if a reasonable reader would so construe the words. Because the objective test of the reasonable reader is to be applied, it is the effect of the text, not the intention of the author, that is assessed.”
I endorse this approach’ (underlining added).
20. It is important to note that there is no requirement for a proven causal link between the hateful expression and actual harm, whether now or pending. In the Court below an arduous endeavour was made in evidence on behalf of the appellant to show just such a link. But that is not what is required to be shown. In paragraph 182 of Qwelane the Constitutional Court explicitly held that: ‘[i]n the context of hate speech prohibitions as civil remedies, a proven causal link between the hateful expression and actual harm is not required.’ This is repeated
in paragraphs 185 – 189.
21. Before applying the law to the facts, it is appropriate to consider briefly the basic approach of the Court below to these legal tests.
THE COURT A QUO’S FINDINGS
22. The court a quo found it unnecessary to deal ‘with each and every issue raised in the papers,’ agreeing with the appellant ‘that the resolution of the first issue is determinative of the remainder.’11 That ‘first issue’ before the court a quo was whether the singing of the two songs – or chants, as they are also described – constituted hate speech as per the Qwelane-reformulated section 10. The structure adopted in the judgment was to set out the procedural history of the matter, then to summarise the evidence of the parties – and in the course of this to reject all the evidence for the appellant as either inadmissible or without probative value or both.
Unfortunately it is only thereafter – from para 12 – that the judgment analyses the regulatory framework, first setting out what the provision (the Qwelane-reformulated section 10) requires, this in para 13. It is respectfully submitted that the High Court fell into fundamental error in adopting the sequence it did. A correct approach would have been first to ask what Qwelane holds, and thereupon to determine that none of the viva voce evidence the Court had allowed both sides to lead was relevant to either of the two inquiries stated above. Any possible relevance lay only with the appellant’s evidence in the single respect to which we revert below.
23. As already indicated it is the statutory provision (in its provisional Qwelane reformulation) which determines fundamentally what the central inquiry entails, and the scope for evidence within it. But, by the time it is considered, not only the admissibility of the evidence but even its probative value, had already been ruled upon.
24. As the court a quo’s findings on the second song are terse, it may conveniently be dealt with first.
The second song – Burn the Boer
25. The court a quo held14 that it had ‘already somewhere in this judgment dealt with the complaint…..[i]t is clear in this regard that Afriforum has failed to make a case that Dr Ndlozi infringed the provision[s] of the Equality Act in singing the impugned song’ (ie the second song, with the lyric, ‘Burn these Boers’).The cross-reference the court evidently had in mind was to para 59 of the judgment (read with paras 53 to 58).15 This is the only place where the second song is dealt with. Par 59 deals with the evidence of the witness Mr Crouse. The court a quo rejected his evidence, because, it said, it considered Mr Crouse not to qualify as an expert, and because if accepted, his evidence anyway had little probative value. The court a quo held that there was simply no evidence (thus whether from Crouse or any other source) ‘to show the harm that arose and may arise from the singing of the second impugned song. Accordingly the complaint stands to fail.’16
26. In short, no proven harm from hate speech, no valid complaint. The court a quo’s judgment regarding the second song is squarely based on what it considered is a requirement of proof by a claimant of actual or potential harm, for a finding of hate speech within the ambit of section of the Equality Act (in its interim formulation by the Constitutional Court, pending amendment) is to be made. Put otherwise, absent proof of harm already caused, or of harm which may arise, the most virulent or inflammatory speech, even though ‘based on’ the prohibited grounds defined in section 1 of the Equality Act, is not hate speech. It may be used with impunity, on the Equality Court’s approach.
27. As pointed out above, the Qwelane formulation does not require proof of either actual or prospective harm arising from the communication which flouts the prohibited grounds. The test in Qwelane is whether the words used themselves could reasonably be construed to demonstrate a clear intention to be harmful or to incite harm and promote or propagate hatred. Proof of consequential harm, actual or potential, or hatred, actual or potential is not required.
28. Although it references Qwelane, the paragraphs in the court a quo’s judgment setting out its findings on the second song show that this is not, with respect, what the court a quo applied in its reasoning. This appears, with respect, to have been the consequence of the court a quo first deciding what evidence to admit and rely on, and only thereafter addressing the requirements of the empowering provision.
The first song: ‘Kill/Kiss the Boer’
29. As to this song or chant, the court a quo (in its ‘Analysis and Evaluation’) held firstly that it had not been shown that the lyrics ‘are based on prohibited grounds set out in the Equality Act’17; secondly, that it was unpersuaded that ‘the lyrics…could reasonably be construed to demonstrate a clear intention to harm or incite to harm and propagate hatred’18; thirdly, that the evidence the appellant advanced – expert or lay – did not assist. The court a quo reasoned further19 that the appellant’s interpretation was ‘literal’. It ‘found no reason to reject’20 Mr Malema’s evidence in which he ‘insisted that the lyrics…should not be interpreted literally but within the context of the struggle and the political message that is sought to be agitated.’ The court a quo also accepted ‘the expert evidence of Prof Gunner’ in which ‘[s]he interpreted the song as not “literally agitating an attack but as a tool to advance the interest of land justice”.’21 The song ‘has to be located within the political context in which Mr Malema is pushing for the land reform and radical economic policy’.22 And so, the court a quo concludes, ‘[I]t is in the current political situation a song directed at articulating the failure of the current government in addressing the issues of economic power, land reform and distribution’.23
THE SONGS CONSTITUTE HATE SPEECH
The songs used words ‘based on’ what are ‘prohibited grounds’
30. The parties allowed the hearing to become dominated by oral and other evidence. That, with respect, is not the place to start. The proper departure point is first to examine what section 10, in its current Qwelane formulation requires. Unfortunately, as already noted, the judgment did not adopt that sequence: it accepted evidence (the respondents’), or rejected it (the appellant’s), in its entirety, then asked what the legal requirements were.24
31. As indicated, at the heart of the post-Qwelane section 10 are two objective inquiries. The torrent of evidence thrust (again, we accept, by both sides) upon the court a quo is determinative of neither. At best there is one limited respect in relation to which the evidence is, secondarily, relevant, to which we revert.
32. As already noted, the judgment records that there is no issue regarding the communication of the words, as alleged by the appellant. What was immediately in issue thereafter was however whether these words were ‘based on’ what are defined in section 1 of the Equality Act as ‘prohibited grounds.’ As noted, this is an objective inquiry where the court must decide whether, in this case, it is based on race, ethnic or social origin.
33. The judgment does not expressly explain why the court a quo considered that reference to ‘Boers’, is not an ethnic reference to South Africans who are Afrikaans-speaking or of Afrikaner descent. Patently, with respect, it is. Every history primer and dictionary so records. The judgment by Lamont J (distinguished and not considered by the court a quo) on a basis addressed below, records this truism.25 A court must have regard to historical facts which are not a matter of contestation.26
34. The court a quo’s approach however appears to have conflated the inquiry as to whether the words used were based on the prohibited grounds with the second part of the reformulated section 10. There, and there only, does intent feature in section 10 – and then to the extent that the words used are required to demonstrate an intent to be harmful or to incite harm and to propagate or promote hatred. If the words used demonstrate these things, it is no answer to say: ‘what I said is not hate speech, because I subjectively never intended that. That may have been what you heard in your ears; it may even have been on my tongue; but it is not what I meant’. But that, in effect, is exactly what the court below upheld.
35. The court a quo’s approach is reflected in its summary of the evidence of Mr Malema. It accepted this testimony in its entirety. In no respect did the Court reject it. The Court thus records that Mr Malema, in relation to the word ‘Boers’ as used in the chant, ‘they [the EFF: see paras  and ] understood that to be directed at the system of oppression’ and ‘to be referring to farmers who
represent the face of land dispossession. The song, need not, according to him, to be taken literally.’27 He himself had been ‘taught not to take the songs in their literal meaning but to understand them to be referring to the oppressive state system’.28
36. But, in neither the first nor the second part of section 10 as reformulated is the inquiry to be answered ‘according to him’. In neither part is it determinative what the speaker says he was ‘taught’ or how ‘to understand them’. ‘[A]ccording to him’ is the Court’s fatal flaw. The Court, as already noted, must answer the first question: do the words used encompass (are ‘based on’) the prohibited grounds? Thus if the speaker refers to moffie or faggot or Sodomite, the question is whether the words are, objectively, ‘based on’ a prohibited ground. The same, by further example, applies if the speaker uses some term which ‘according to him’ is not anti-semitic. If the court holds that it is based on an ethnic denigration of Jews, the first part of section 10 is answered. So too if the target is another ethnic group, or grouping of groups, whether the message is encoded (as it was in Rwanda) as ‘cockroaches’ or in South Africa, nkwerekwere.
37. It is therefore submitted that both songs’ words were based in material part on a prohibited ground. The reference to ‘Boere’ was explicitly ethnic,29 as much as Armenian would be in 1915 or Tutsi in 1997. Mr Malema’s assertion – the Court refers to it, twice, as his ‘insistence’30 regarding how he or others allegedly saw it – is irrelevant. ‘[A]ccording to him’ is thus a repeat-error. So too is the court’s acceptance of Prof Gunner’s elaborately argued (and elaborately metaphorical) interpretation. What the court should have held is that ‘[b]ecause it has long been held that an expert may not usurp the adjudicative functions of our courts, the experts in this matter could not be used to determine the meaning of the statements.’31
The words also demonstrate an intent to incite harm or propagate hatred
38. As the Constitutional Court confirmed in Qwelane32, and as already noted above, the subjective intent of the author is thus irrelevant. The test is objective. If a reasonable person reading the text could understand it to mean an incitement to cause harm, the test is met. The court a quo with respect correctly quoted the Qwelane test – but having already accepted Mr Malema’s evidence as to what he ‘intended’ and had been ‘taught’, it did not apply it.
In paragraph 93, the court a quo correctly found, in reliance on Sutherland J’s test in Khumalo33, that the test for hate speech is objective, whether the utterance ‘could be reasonably construed to demonstrate a clear intention to incite harm’. But by then it had already accepted in toto the evidence of Mr Malema and Prof Gunner (with their ‘insistence’ on what the words meant).
39. The court a quo, with respect, also wrongly held in this regard that in some manner the appellant had to ‘prove’ what the words, reasonably construed, convey to their target audience. Either the words do the work, or they do not. Here they clearly did: ‘kiss’ was, as the evidence shows, an artful substitute. As with Mark Antony’s oration in Julius Caesar (‘And sure, Cassius is an honourable man’34) the exactly inverted meaning is clear to the knowing audience. Again the inconsistency with the court a quo’s explicit adoption in para  of the wholly objective Khumalo test.
40. This Court’s findings in Hotz35 are pertinent. During the Rhodes Must Fall protests a person wore a t-shirt with the words ‘KILL ALL WHITES’ painted on it. He contended that it had a small ‘s’ painted in front of the word kill which meant it actually read ‘sKILL ALL WHITES’. This court found that the reaction by people who saw it, and who would not dare venture close enough to see an imperceptible letter ‘s’, would be to take it at ‘face value’ as ‘an incitement to violence against white people…that all white people should be killed.’ This Court continued by finding that ‘the fact that [he] sought to explain away the slogan and suggest that it said something other than what it clearly appeared to say, is itself a clear indication that he recognised its racist and hostile nature.’
The same in this matter. People stand to react to ‘Kill the Boer’ as an incitement to violence against Boers – a call to kill them. Mr Malema knows this – that is why he substitutes kiss for kill. It is plain from the wider context, notably the accompanying gestures of pointing and firing a gun, that Mr Malema is not exhorting osculatory affection. (The hand gestures mimicked shooting, not blowing a kiss. How the court below could have left the potent gestures out of account entirely is not apparent.)
41. It is in this regard that the court a quo stands, too, to be held to have erred in its holus bolus rejection of the appellant’s evidence. The evidence of the two lay witnesses affirmed the social milieu in which the words were used and their impact. So too the study, and book, by Mr Roets36. The evidence showed a Qwelane-section 10 inquiries are objective. On these matters neither Mr Roets nor Mr Malema nor any society traumatised by its past and by a high level of violence on farms, as elsewhere. In this single respect, it is submitted, that part of the viva voce evidence was relevant and admissible. The appellant’s case however stands whether or not all its evidence was correctly disallowed. It is for this reason, and to lessen both the burden on this court and the incurrence of costs, that the appellant proposed (to no avail) that the viva voce evidence be dispensed with37.
42. Significantly, in their heads of argument before the court a quo the respondents conceded that some people might take ‘Kill the Boer’ literally. Their case is that the test for hate speech has not been met because there is no proof that particular killings have been directly consequential to particular exhortations, such as those which are common cause in this case, to ‘kill/kiss’ or to ‘burn’.38 Apart from showing that the respondents have always argued that proof of a link between incitement and harm was required,39 this concession is fatal to their case. It is itself an admission that the song could reasonably be construed to demonstrate a clear intention to incite harm.40
Taking the gestures with the gun into account
43. As noted, the court found as a fact that not mere words were involved in the second song. Mr Malema used the gesture of a gun. The gesturing underscored the verbal message – and showed up the ‘kiss’ paraphrase for what it was: as Lamont J has held41, a coded message for kill, much like Mark Antony’s to the crowd about Cassius.
44. Lamont J did a careful historical and contextual analysis of ‘Dubula ibhunu/ Kill the Boer’. He found that the audience of the song – as sung at political rallies – is the whole public, especially White Afrikaners as the target group. It makes no difference if the message is encoded differently. On facts identical to this matter he found it sufficient that there is ‘a variety of members of society who act for large constituencies and who say that their constituencies are affected in that they perceive the song to be harmful and/or hurtful towards them.’ Lamont J finds that this reaction is due ‘to the context and manner in which Malema repeatedly sang the song and exploited the publicity his singing the song had in translation as well as in the original language.’42
45. It is submitted that on the authorities just cited and not addressed by the court a quo, an accompanying gesture of a gun (or knife or rope or whip) is deeply contextual to the actual words used.43 Also that such a gesture could – all that the provision requires – reasonably be interpreted as reflecting an intent to be harmful, to incite harm, and to propagate hatred. Like drawing a finger across the throat, or swinging a rope, the gesture may even add such blatant specificity as the accompanying words may lack.
Proof of harm?
46. As noted above, proving incitement to cause harm does not, in turn, require proof of either actual harm arising or that potential harm may arise. Incitement is long- established in law as just that: instigating others to do something. There is, with respect, no basis to import into section 10, as reformulated by the highest court, a provision which is not there, and which is contrary to the developed law as to what constitutes incitement.
47. In answering the question whether a message could have the effect of inciting harm or constitute incitement to cause harm in the eyes of the reasonable reader Sutherland J in Khumalo gives a wide definition of harm. This goes further than merely physical harm, to inter-racial hostility in the form of reactions ranging from those who share those views to those in the target group, to harm to social cohesion in South Africa.44
48. Lamont J also refers to this risk of harm pointing out that all hate speech has an effect, not only on the target group but also on the group partaking in the utterance. He states that such a ‘group and its members participate in morally corrupt activity which detracts from their own dignity. It lowers them in the eyes of right-minded balanced members of society who then perceive them to be social wrongdoers.’45
49. Lamont J states further that ‘the words are inflammatory’ and ‘[a]ll conduct by more than one person has as its source the words of at least one person. It is the words of one person motivating others that leads to action by those persons. All genocide begins with simple exhortations which snowball. Words provide the stimulus for action, the means to numb the natural repugnance against hurting humans and the reward which is to be harvested after action. Words are powerful weapons which if they are allowed to be used indiscriminately can lead to extreme and unacceptable action.’46
50. In a similar vein Sutherland J referred to the ‘risk of spiralling invective with uncertain but frightening possibilities’.47 Simple exhortations can numb a natural repugnance to hurt humans and easily snowball. It is this that makes Mr Malema’s comments in cross-examination so ominous. He said ‘when’ – not if – the time to attack comes he will not sing. He will simply give the command. This may include slaughter, on grounds of race.48
51. The issue is whether speech is protected. It is not protected when it enjoins hate.
Section 16 of the Constitution protects freedom of expression but not ‘advocacy of hatred based on race, ethnicity, gender or religion and that constitutes incitement to cause harm’. As held by Langa DCJ in Islamic Unity49 ‘[s]ection 16(2) defines the boundaries beyond which the right to freedom of expression does not extend’.50 As stated by Sutherland J in Gordhan51 the scope of section 10 itself is circumscribed by section 16(2)(c).
52. In reliance on Afri-Forum I and Masuku, Sutherland J in Khumalo52 held that a significant aspect of the conduct prohibited by section 10 is that, but for the exceptions set out in section 12, there can be no defence or escape from liability. Justification and fair comment offer no defence. If a person uttered the impugned words, even if it was not their intention, they are liable.
53. In this regard the extended conversation at the end of his evidence – stretching to thirteen transcribed pages – between Mr Malema and the court a quo is insightful.53 On pages 112 line 10 – page 113 line 9 in particular the court asks him what his views would be supposing there are non-racist farmers committed to the new dispensation who are scared by the song.
We repeat that the test is objective and not to be determined according to Mr Malema’s views on the matter. Compare nevertheless his completely contradictory response. In answer to the court’s question he says ‘[they] will have to be sensitive’ to such farmers and ‘[would] tell them that they have no reason to be scared’. Earlier he repeated how utterly unmoved he was to the plight of Mr Prinsloo, a fluent isiZulu-speaking veterinary surgeon, who lost his wife to death, his daughter to emigration and his ability to walk due to a farm attack.54
54. Once a court, after applying the Qwelane section 10 test, finds that utterances constitute hate speech it automatically follows that any section 16 protection falls away. The court a quo however found in para 112 that ‘a reasonable listener, would conclude that the song does not constitute hate speech but rather that it deserves to be protected under the rubric freedom of speech’. This was with respect a misdirection and a conflation of tests. The test is not whether a reasonable reader would conclude that the song deserves to be protected as freedom of speech. Hate speech is never protected.
55. This court in Hotz55 illustrated the correct approach when it found that the slogan painted on the t-shirt ‘was advocacy of hatred based on race alone and it constituted incitement to harm whites. It was not speech protected by s 16(1) of the Constitution.’
Distinguishing Afri-Forum I (per Lamont J)
56. All that changed with Qwelane is that ‘hurtful’ words no longer constitute hate speech. The attempt to distinguish Lamont J’s judgment with respect disregards that it was not based only on what he considered (on the law as it then stood) was ‘hurtful’: he expressly found that the words incited harm and propagated hatred. It is also no basis for distinction, with respect, that by agreement Lamont J’s order was substituted. His judgment, comprising an analysis of the law, also cannot be expunged by agreement between the parties to substitute his order. The difference between orders and judgments was laid down in Ursula Mansions56, and recently in Brummer.57
57. The court a quo was with respect bound by its earlier judgment between the same parties on the same issues, which the deletion by judicial amendment of ‘hurtful’ did not change. The court a quo could depart from the Lamont J judgment only if it held it to be clearly wrong. It did not do so. Accordingly the judgment qua precedent stands.
58. The court a quo, with respect, accordingly misdirected itself in evidently considering that an order substituted by agreement deletes the judgment. It, in any event, with respect, conflated the effect of any judgment inter partes with its status as a legal precedent. The pronouncements of law by Lamont J (as a court of the same status) stood, save to the extent they were held to be plainly wrong. No such findings were made.
Misdirections regarding other judgments
59. The court a quo, with respect, further relied on an entirely incorrect paraphrasing of Sutherland J (as he then was) in Khumalo58 by Molefe J in Sanef v EFF59. In the ‘other words’ of Molefe J, what Sutherland J meant is that ‘even if the prohibited utterances in question could qualify as hate speech on its [sic] terms but fail to incite, or reasonably construed as inciting harm, no liability could arise in respect of section 10.’
60. Unfortunately, the Court a quo with respect took this at face value. In fact, Molefe J was demonstrably in error. Citing what Sutherland J had in fact held would have revealed this. The court a quo however looked no further, and interpreted Molefe J’s mistake regarding what Sutherland J had actually held to mean that actual harm or incitement was required and failed to distinguish between intention and actual occurrence.
61. What Sutherland J in reality said in paragraph 88, was ‘[t]he test for hate speech is whether Khumalo’s utterances ‘….could be reasonably construed to demonstrate a clear intention to ‘incite harm’. This postulates a reasonable reader. It asks what a reasonable reader could think about the speech. If a reasonable person reading the text could understand it to mean an incitement to
cause harm, the test is met. What is assessed by the reasonable reader is the effect of the utterances on readers in general. The word “could” in section 10(1) must be emphasised too, because the perspectives of ‘reasonableness’ inevitably are located within a band of reasonableness, and the test corresponds to that reality…The standard of the reasonable person, applied to section 10(1), means, therefore, whether a reasonable person could conclude (not inevitably should conclude) that the words mean the author had a clear intention to bring about the prohibited consequences’ (underlining added).
62. The court a quo, despite invoking an objective test, with respect in reality applied a subjective test. It simply preferred Mr Malema’s explanation of what he would like (at least, on one of his versions) the words to mean. It did this even before it had analysed what the legal requirements of the Qwelane-section 10 are. And then the judgment circles back to re-assert a subjective approach, after having disavowed it, invoking Khumalo. The test is however objective and if a reasonable person hearing it could understand it to mean an incitement to cause harm the test is met. This case plainly meets the test and does away with the second error of the court a quo (erroneously adopting in this respect a paraphrase of Khumalo), which was to require actual or potential incitement to harm.
63. As Lamont J stated in Afri-Forum I, times have changed. Whether singing ‘Kill the Boer’ may have been acceptable while the enemy, the regime, remained the enemy of the singer, is not the issue now. Pursuant to the agreement contained in the Constitution ‘the enemy has become the friend, the brother’60 and the song should no longer be sung. As Sutherland J stated in Khumalo61: ‘[i]n South Africa…our policy choice is that utterances that have the effect of inciting people to cause harm [are] intolerable because of the social damage [they] wreak and the effect [they have] on impeding a drive towards non-racialism.’
64. It is submitted that the court a quo erred in its finding and the appeal should be upheld. The respondents’ singing in public of the songs known as Dubula ibhunu (Kill/Kiss the Boer) and Bizani iFire Brigade (containing the exhortation to Burn these Boers) constitute hate speech in terms of section 10(1) of the Equality Act.
65. The terms of the proposed order are attached. It is restricted to the public use by the respondents of the two songs containing clear exhortations to kill or burn, directed at Boere/farmers/ibhunu.
66. The respondents, as the attached correspondence establishes, were specifically asked (in terms of SCA Rule 8(8)(d)) to cooperate in not burdening this court with the full record below, avoiding, too, the burden of unnecessary costs on the parties. They declined. Given that the respondents’ conduct in this regard justifies a special order as to costs, because it meets the unreasonableness threshold test, long applied by our courts,62 and spelt out in the rule, the appellant asks that the costs unnecessarily incurred in this court despite the proposal properly made in terms of should be for the respondents’ account.
JEREMY GAUNTLETT SC KC
MICHAEL TSELE CHRISTOPHER BURKE
Chambers, Cape Town Wednesday, 2 August 2023
ANNEXURE A: DRAFT ORDER
1. The order of the High Court is set aside and replaced with the following order:
‘1. The words and translations of words, phrases and songs set out below constitute hate speech:
dubula amabhunu baya raypha
shoot to kill. Kiss the Boer – the farmer Shisa lamabhunu
2. The respondents are interdicted and restrained from any public use, singing or chanting of the words, phrases or songs set out in paragraph 1.
3. The respondents are to pay the costs of the appellant, jointly and severally, the one paying the others to be absolved, excluding costs incurred in the court a quo relating to the calling of the four witnesses by the appellant.
4. The respondents shall pay the costs of the appeal, jointly and severally, the one paying the others to be absolved, including the costs of the application for leave to appeal, such costs to include the costs of three counsel.
5. As regards the costs of preparation of the record on appeal, the respondents are directed, in terms of Rule 8(8)(d), to pay the costs on the scale as between attorney and client.
TABLE OF AUTHORITIES
1. Constitution of South Africa, 1996
2. Equality Act 4 of 2000
3. *Afri-Forum v Malema  ZAEQC 2; 2011 (6) SA 240 (EqC)
4. Azanian Peoples Organisation (AZAPO) and Others v President of the Republic of South Africa and Others  ZACC 16; 1996 (4) SA 672 (CC)
5. Community of Grootkraal v Botha NO and Others  ZASCA 158; 2019 (2)
SA 128 (SCA)
6. Democratic Alliance v Brummer  ZASCA 151
7. Gordhan v Malema  ZAEQC 5; 2020 (1) SA 587 (GJ)
8. Hotz v University of Cape Town  ZASCA 159; 2017 (2) SA 485 (SCA)
9. Islamic Unity Convention v Independent Broadcasting Authority  ZACC 3; 2002 (4) SA 294
10. *Qwelane v South African Human Rights Commission and Another  ZACC 22; 2021 (6) SA 579 (CC)
11. *South African Human Rights Commission v Khumalo  ZAGPJHC 528; 2019 (1) SA 289 (GJ)
12. South African Human Rights Commission obo South African Jewish Board of Deputies v Masuku  ZACC 5; 2022 (4) SA 1 (CC)
13. South African National Editors’ Forum and Others v Economic Freedom Fighters and Another  ZAEQC 6
14. Western Johannesburg Rent Board and Another v Ursula Mansions (Pty) Ltd 1948 (3) SA 353 (A)
End of 2016 – October 2020
The impugned songs are sung or chanted by the Respondents at various events.
Core bundle pages 15 – 18
7 December 2020
Core bundle page 1
25 August 2022
High Court judgment handed down
Record (vol 11) page 1951
4 October 2022
High Court grants leave to appeal
Record (vol 11) page 1966
1 To which we revert in paragraph 16 below.
2 See judgment and order – record (vol 11) pages 1910 – 1951 & 1952.
3 Judgment (vol 11) page 1936.
4 Core bundle pages 15 – 17, paras 26 – 37 (FA) & transcriptions of recordings, core bundle pages 31 – 52. Not materially denied in the pleadings: Mr Malema implies that he sang ‘kiss’ instead of ‘kill’ and that he was, in any event, not precluded by court order from singing either of the words – core bundle page 83 para 96 – page 85 para 103 (RA). It was admitted by Mr Malema in cross-examination where he explains what he sang: ‘Shoot to kill. Kill a man’, ‘kiss the Boer, kiss the farmer’ – (vol 8) page 1534 lines 5 & 22. ‘The next part of the chant has you firing guns. And saying pow-pow. Is that correct?’ ‘Ja.’ – Record (vol 8) page 1549 line 25 – page 1550 line 2.
5 Record (vol 11) page 1945, para 100 (Judgment).
6 Record (vol 11) page 1934, para 70 ad finem (Judgment).
7 Core bundle pages 17 – 18, paras 38 – 41 (FA) & core bundle page 86, para 110 (RA) ‘I can therefore not definitively confirm that the specific words were sung by an EFF member/s though it is probable that some of the EFF members also participated in the general singing of the song in question.
8 Core bundle page 17, para 41 (FA) & record (vol 1) page 87, para 114 (RA).
9 Qwelane v South African Human Rights Commission  ZACC 22; 2021 (6) SA 579 (CC).
10 Quoted in para  of the Court a quo’s judgment, record (vol 11) page 1941.
11 Record (vol 11) page 1950, para 114 (Judgment).
12 Record (vol 11) page 1938, para 83 (Judgment).
14 Record (vol 11) page 1945, para 99 (Judgment).
15 Record (vol 11) page 1929, para 53 – page 1931, para 59 (Judgment).
16 Record (vol 11) page 1931, para Id para 59 ad finem (Judgment).
17 Record (vol 11) page 1946, para 101 (Judgment).
19 Record (vol 11) page 1946, para 103 – page 1948, para 110 (Judgment).
20 Record (vol 11) page 1946, para 104 ad finem (Judgment).
21 Record (vol 11) page 1947, para 105 (Judgment).
22 Record (vol 11) page 1946, para 109 (Judgment).
23 Record (vol 11) page 1949, para 111 (Judgment).
24 “’Let the jury consider their verdict,’ the King said, for about the twentieth time that day. ‘No, no!’ said the Queen. ‘Sentence first – verdict afterwards’. Lewis Carroll, Alice’s Adventures in Wonderland (1865) ch XVII.
25 Afri-Forum v Malema  ZAEQC 2; 2011 (6) SA 240 (EqC) paras 4, 5 & 105 (‘Afri-Forum I’) – ‘The regime was destroyed at the time of the transformation of the country into a democracy. It is no more. Post democracy the song was none the less sung, seeking its destruction. The response of Malema to this conundrum was to say that the regime lives on in the form of the untransformed person who holds benefits conferred upon him by the regime and which he has not relinquished. He accepts that there is an object to the verb and that that object is alive and well and living in South Africa. It is a simple matter to identify the object. It is those persons who received benefit from and who promoted the regime. These persons are, broadly speaking, the white Afrikaans speaking members of society.’
26 Community of Grootkraal v Botha NO and Others  ZASCA 158; 2019 (2) SA 128 (SCA) para 21, and authorities collected in fn 27.
27 Record (vol 11) page 1935, para 71 (Judgment).
28 Record (vol 11) page 1934, para 70 (Judgment).
29 Afri-Forum I note 24 above paras 59 – 61, where Lamont J explains the double meaning of the phrase ‘Dubula Ibhunu’ with its literal meaning being ‘shoot the Boer’.
30 Record (vol 11) page 1946, para 104 (Judgment).
31 South African Human Rights Commission obo South African Jewish Board of Deputies v Masuku ZACC 5; 2022 (4) SA 1 (CC) para 145.
32 Qwelane note 9 above at para 97.
33 South African Human Rights Commission v Khumalo  ZAGPJHC 528; 2019 (1) SA 289 (GJ).
34 William Shakespeare, Julius Caesar, Act III Scene II.
35 Hotz v University of Cape Town  ZASCA 159; 2017 (2) SA 485 (SCA) para 68.
36 His qualification as an expert did not lie only in the fact of his postgraduate degree cum laude in constitutional law: he had made a three-year study of the subject of farm attacks. This particular knowledge qualified him to give opinion evidence as an expert. He was not disqualified as an expert because he is employed by the Appellant. The law reports are studded with skilled employees called as expert witnesses by their own employer. Of course the probative value may, in certain circumstances be lessened by the fact.
All this, however, it is submitted, is a side-issue, given the fact that, as demonstrated, the two essential Qwelane-section 10 inquiries are objective. On these matters neither Mr Roets nor Mr Malema nor any other witness was entitled to, in the Court’s word, ‘insist’ on the conclusion.
37 See paragraph 16 above.
38 Respondent’s heads of argument in court a quo paras 204-206, 213 & 214.
39 Leading to arguments such as that because one of the witnesses Mrs Muller, who survived a farm attack, was able to process what happened to her, she suffers (suffered?) no trauma. See in this regard Respondent’s heads of argument in court a quo paras 166 & 238.
40 For the binding nature of a legal concession correctly made in heads of argument see para 16 of Azanian Peoples Organisation (AZAPO) and Others v President of the Republic of South Africa and Others  ZACC 16; 1996 (4) SA 672 (CC).
41 Afri-Forum I note 25 above para 39: ‘Those gestures form part of the context and will be relevant to determining the reasonable construction to be placed upon the words. See for example Phillips v DPP 2002 (5) SA 555 (W) at para 14-17; Botha Eiendomme (Edms) Bpk v Ekple-Epoh 2000 (4) SA 466 at
471 para 3.3 and S v Seshama 1991 (2) SA 860 (SCA) at 879.)’.
42 Id paras 91, 93 and 95.
43 Afri-Forum I note 25 above para 104 ‘When the gestures made by Malema are added to the context then it is clear that the words concern the use of a weapon – a gun. Whether the verb alone means destroy or shoot makes no difference. The verb contains an exhortation to violence. The gesture imports the weapon. Hence the mechanism by which the exhortation is to be implemented is by the use of the weapon, a gun. In reaching this conclusion, I am conscious that there are many ways by which destruction can take place, shooting is but one of them. In the context of the song the gesture provided the limitation on the words. The person to be shot is the object of the verb namely the regime. The regime included the Boere or white Afrikaans speaking sector of society. This sector might also include farmers.’
44 Khumalo note 33 above paras 92 – 97.
45 Afri-Forum I note 25 above para 94.
47 Khumalo note 33 above para 102.
48 Evidence of Mr Malema, Transcript – 16 February 2022 – record (vol 9) page 1640 lines 17 – 24; See similarly Mr Malema’s admission that he said in a speech that ‘We are not calling for the slaughter of white people, at least for now’ and his refusal to pledge that he would not at some future date call for the slaughter of white people. Evidence of Mr Malema, Transcript – 17 February 2022 – (vol 9) page 1654 line 13 – page 1656 line 25.
49 Islamic Unity Convention v Independent Broadcasting Authority  ZACC 3; 2002 (4) SA 294 (CC).
50 Id para 32.
51 Gordhan v Malema  ZAEQC 5; 2020 (1) SA 587 (GJ).
52 Khumalo note 33 above para 98.
53 Evidence of Mr Malema, Transcript – 17 February 2022 – record (vol 9) page 1715 line 24 – page 1727 line 1.
54 Evidence of Mr Malema, Transcript 17 February 2022, (vol 9) page 1707 line 24 – page 1708 line 4. ‘I said I am not moved…I said I am not moved. Let me repeat five times I am not moved by that case you brought here. I am not moved. And if that will make me lose a case, let me lose it. I am not moved.’
55 Hotz note 35 above para 68.
56 Western Johannesburg Rent Board and Another v Ursula Mansions (Pty) Ltd 1948 (3) SA 353 (A).
57 Democratic Alliance v Brummer  ZASCA 151 para 39.
58 Khumalo Note 33 above para 88.
59 South African National Editors’ Forum and Others v Economic Freedom Fighters and Another  ZAEQC 6 para 36.
60 Afri-Forum I note 25 above para 108.
61 Khumalo note 33 above para 102.
62 At least since Re Alluvial Creek 1929 CPD 532.
Photos: Putin and the JEWS: The false belief that Putin will save the White Race
In 2017, in this article I pointed out how much nonsense is being written in the White Right about Putin being some kind of saviour of the White Race. I pointed out his links to Jews. Putin is a Jew-friendly Russian ruler. Since then I‘ve discovered even more data to support this.