Association for Mineworkers and Construction Union and Others v Anglo Gold Ashanti Limited and Others.

CONSTITUTIONAL COURT OF SOUTH AFRICA In the matter between: ASSOCIATION OF MINEWORKERS AND CONSTRUCTION UNION EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J420/19) EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J424/19) INDIVIDUALS LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J430/19) THE PERSONS WHOSE NAMES APPEAR ON ANNEXURE “A1” TO THE NOTICE OF MOTION (J431/19) EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J432/19) EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J438/19) EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J439/19) EMPLOYEES LISTED IN ANNEXURE “A” TO THE NOTICE OF MOTION (J440/19) MEMBERS OF AMCU EMPLOYED BY THE APPLICANT (J442/19) Case CCT 233/20 First Applicant Second Applicant Third Applicant Fourth Applicant Fifth Applicant Sixth Applicant Seventh Applicant Eighth Applicant Ninth Applicant Tenth Applicant MEMBERS OF AMCU EMPLOYED BY THE APPLICANT (J444/19) Eleventh Applicant and ANGLO GOLD ASHANTI LIMITED T/A ANGLO GOLD ASHANTI (J420/19) First Respondent LONMIN PLATINUM COMPRISING WESTERN PLATINUM AND EASTERN PLATINUM LIMITED T/A LONMIN (J424/19) Second Respondent RUSTENBURG PLATINUM MINES LIMITED T/A RUSTENBURG PLATINUM MINES (J430/19) Third Respondent HARMONY GOLD MINING COMPANY LIMITED T/A HARMONY GOLD (J431/19) VILLAGE MAIN REEF (PTY) LIMITED, TAU LEKOA (PTY) LIMITED AND KOPANONG (PTY) LIMITED T/A VILLAGE MAIN REEF (J432/19) NORTHAM PLATINUM LIMITED T/A NORTHAM PLATINUM (J440/19) MARULA PLATINUM (PTY) LIMITED (J439/19) IMPALA PLATINUM LIMITED T/A IMPALA PLATINUM (J440/19) GLENCORE OPERATIONS SA (PTY) LIMITED (J443/19) BUSHVELD VAMETCO ALLOYS (PTY) LIMITED (J444/19) Fourth Respondent Fifth Respondent Sixth Respondent Seventh Respondent Eighth Respondent Ninth Respondent Tenth Respondent Neutral citation: Association of Mineworkers and Construction Union and Others v Anglo Gold Ashanti Limited t/a Anglo Gold Ashanti and Others [2021] ZACC 42 Coram: Khampepe ADCJ, Jafta J, Madlanga J, Majiedt J, Mhlantla J, Pillay AJ, Theron J, Tlaletsi AJ and Tshiqi J. Judgments: Pillay AJ (majority): [1] to [111] Jafta J (concurring in costs orders only): [112] to [160] Theron J (concurring): [161] to [170] Heard on: 6 May 2021 Decided on: 12 November 2021 Summary: Section 66(2)(c) of the Labour Relations Act 66 of 1995 — lawfulness of secondary strikes — section 66(2) imports a proportionality assessment Secondary employer — interdicting secondary strikes — proportionality assessment ORDER On appeal from the Labour Appeal Court, the following order is made: (a) (b) Leave to appeal is granted. Save as set out below, the appeal against the judgments of the Labour Court and the Labour Appeal Court is dismissed. (c) The costs orders in the Labour Court and the Labour Appeal Court are set aside. (d) Each party shall pay its own costs in the Labour Court, the Labour Appeal Court and this Court. JUDGMENT PILLAY AJ (Khampepe ADCJ, Madlanga J, Majiedt J, Mhlantla J, Tlaletsi AJ and Tshiqi J concurring): Introduction “[The right to strike] is to the process of bargaining what an engine is to a motor vehicle.”1 [1] The Labour Court interdicted the intended secondary strikes at 10 mining companies in support of a primary strike at Sibanye Gold Limited t/a Sibanye Stillwater (Sibanye).2 That primary strike has long been resolved, and consequently, the dispute is no longer live. However, this application for leave to appeal poses a discrete question of law: does section 66(2)(c) of the Labour Relations Act3 (LRA) import the principle of proportionality when assessing the substantive lawfulness of secondary strikes? It is to address this question, that I pen this judgment notwithstanding the absence of a live dispute. Parties [2] The first applicant is the Association of Mineworkers and Construction Union (AMCU). It brings this application together with its members, the second to eleventh applicants, who are employees of the first to tenth respondents. The first to eleventh 1 National Union of Metal Workers of SA v Bader Bop (Pty) Ltd [2002] ZACC 30; 2003 (3) SA 513 (CC); 2003 (2) BCLR 182 (CC) (Bader Bop) at para 67. 2 Anglo Gold Ashanti Limited t/a Anglo Gold Ashanti v Association for Mineworkers and Construction Union (2019) 40 ILJ 1552 (LC) (Labour Court judgment) at paras 273-5 and 278. The order of the Labour Court was upheld on appeal in Association of Mineworkers and Construction Union v Anglogold Ashanti Limited t/a Anglogold Ashanti [2020] ZALAC 45; (2020) 41 ILJ 2763 (LAC) (Labour Appeal Court judgment). 3 66 of 1995. 4 PILLAY AJ applicants will be referred to as either “AMCU” or “the applicants” and together they represent all AMCU members at various companies who intended to participate in the secondary strikes. [3] The first to tenth respondents are Anglo Gold Ashanti Limited t/a Anglo Gold Ashanti; Lonmin Platinum Comprising Western Platinum Limited and Eastern Platinum Limited t/a Lonmin; Rustenburg Platinum Mines Limited t/a Rustenburg Platinum Mines; Harmony Gold Mining Company Limited t/a Harmony Gold; Village Main Reef (Pty) Limited; Tau Lekoa (Pty) Limited and Kopanong (Pty) Limited t/a Village Main Reef; Northam Platinum Limited t/a Northam Platinum; Marula Platinum (Pty) Limited; Impala Platinum Limited t/a Impala Platinum; Glencore Operations SA (Pty) Limited and Bushveld Vametco Alloys (Pty) Limited. They are mining companies operating in the Republic of South Africa. Collectively, they will be referred to as “the respondents” or “the secondary employers” as the context requires. Only the second, third, fourth, sixth, seventh, eighth and tenth respondents participated in this appeal. Factual background [4] Sibanye is a public company operating in the gold and platinum mining industry. It is the primary employer in this dispute concerning secondary strikes. Wage negotiations with AMCU commenced before the Minerals Council of South Africa (Minerals Council) on 11 July 2018. Shortly thereafter, AMCU referred a mutual interest dispute about wages and other conditions of employment to the CCMA. A certificate of non-resolution of the dispute was issued on 26 September 2018, resulting in AMCU commencing the strike against Sibanye on 21 November 2018. By 20 February 2019, the primary strike was still under way. AMCU gave notice to the respondents of its intention to commence secondary strikes for a week, beginning on 28 February 2019, and concluding on 7 March 2019. [5] All AMCU members employed by the respondents were to be involved in the secondary strikes. AMCU’s representivity varied amongst the secondary employers. 5 PILLAY AJ For example, at four respondents, AMCU held a sizeable majority representation. The respondents employed members of other trade unions too. [6] The secondary strikes would have severely disrupted the respondents’ operations and prevented production for their entire duration. The respondents, miners of mainly gold and platinum, stood to lose millions of Rands in revenue. Formidable costs would have been incurred in shutting down and then restarting the mining operations as well as in maintenance services to keep mines continuously safe. [7] Although the seven-day multi-employer secondary strikes would have constituted only 2% of the working year, the daily losses each respondent would have experienced was estimated to exceed R11 million, and as much as R214 million in one instance. Cumulatively, the losses over seven days across all ten respondents would exceed R2 billion. That is not all. The practicalities and safety considerations of a total shut down for the respondents and non-striking employees militated against allowing the secondary strikes to proceed. [8] The standard “no work, no pay” principle applied to striking employees. That, and hard bargaining, had the potential of exacerbating the level of violence associated with the primary strike. Already, by then, seven non-striking employees at Sibanye had died. [9] The respondents’ stance was that the secondary strikes would have no effect on the business of Sibanye, but would cause their businesses significant harm. Taking the opposite view on the effect of the secondary strikes, AMCU acknowledged the harmful impact of the secondary strikes on the respondents, but dismissed it as par for the course. This dispute whittles to the interpretation and application of section 66(2)(c), which provides – 6 “No person may take part in a secondary strike unless the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer”.4 PILLAY AJ [10] The respondents were of the view that the secondary strikes would have been unreasonable in terms of the above provision. They turned to the Labour Court to stop the secondary strikes. Litigation History Labour Court [11] Each respondent raised distinct arguments about the harmful effects that the secondary strike at its workplace would have on its operations without having any direct or indirect effect on collective bargaining between AMCU and the primary employer. The gist of the respondents’ submissions was that the secondary strikes were disproportionate to the primary strike. Therefore, the secondary strikes were not reasonable in relation to their possible direct or indirect effect on the business of the primary employer. The respondents contended that no nexus existed between Sibanye and themselves and that they were unable to exert any influence over the outcome of the primary strike, because exerting pressure was not possible when they shared no commodities, assets or resources with Sibanye. [12] Moreover, the respondents continued, the cause, for which each secondary strike was called, was not one in which the employees of the secondary employers had 4 Section 66(2) states that no person may take part in a secondary strike unless— “48 hours’ notice of the application has been given to the respondent: However, the Court may permit a shorter period of notice if— (a) (b) (c) the applicant has given written notice to the respondent of the applicant’s intention to apply for the granting of an order; the respondent has been given a reasonable opportunity to be heard before a decision concerning that application is taken; and the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer.” 7 PILLAY AJ any interest or expectations of gain. And worker solidarity alone was not enough to justify a secondary strike. The respondents argued further that exposing their non-union employees to the risk of violence, when they had no control or influence over the primary strike, would be grossly unfair. Therefore, the respondents averred, the secondary strikes did not meet the requirements set out in section 66(2)(c). Consequently, the secondary strikes were unprotected and fell to be interdicted. [13] Opposing the application for the interdict, AMCU argued that the right to strike in section 23(2)(c) of the Constitution5 does not distinguish between the right to participate in a primary and a secondary strike. It submitted that the proposed secondary strikes complied with section 66(2), and that the nature and extent of the secondary strikes were reasonable in relation to the possible direct or indirect effect that the secondary strikes might have on the business of the primary employer. The intended secondary strikes would be for the short period of seven days, or 2% of the working year. Furthermore, they would take the form of a stay-away which would alleviate the risk of violence. [14] AMCU’s primary argument turned on the role of the Minerals Council. As a registered employers’ organisation, it could influence collective bargaining in the mining industry. AMCU banked on the secondary employers, as members of the Minerals Council, bringing their influence to bear on Sibanye’s Chief Executive Officer, Mr Neal Froneman, the Vice President of the Minerals Council. AMCU reasoned that the secondary employers could swing sentiment favourably towards AMCU’s members at Sibanye to avert the costs and inconveniences of secondary strikes in their own workplaces. 5 Section 23(2) of the Constitution states: “Every worker has the right— (a) (b) (c) to form and join a trade union; to participate in the activities and programmes of a trade union; and to strike.” 8 PILLAY AJ [15] All in all, continued AMCU, the secondary strikes would be eminently reasonable. Economic loss was not a factor that ranked highly when considering industrial action. Instead, economic loss was an inevitable consequence which underpinned the purpose of industrial action. Even before the secondary strikes commenced, ratings agency Moody’s announced its intention to downgrade Sibanye’s credit rating and share price. This showed, so the argument went, that the simultaneous secondary strikes would fulfil their legally protected purpose by inducing economic pressure on Sibanye. [16] In reply, the respondents delivered an affidavit to which they annexed a signed statement from Sibanye’s Chief Executive Officer stating that Sibanye remained resolutely unmoved by the secondary strikes. The Minerals Council also weighed in to deny having any influence on the business of the primary employer and collective bargaining between AMCU and Sibanye. Instead, its testimony was that its primary function was to lobby and advocate on behalf of its members on policy matters. [17] The Labour Court accepted that the procedural requirements in section 66(2)(a) and (b) had been fulfilled.6 The application to declare the strike unprotected, therefore, turned only on the requirement of reasonableness in section 66(2)(c).7 [18] The Labour Court held that section 66(2)(c) required a consideration of the direct or indirect effect of the secondary strikes on the primary employer while also entailing an enquiry into the impact on the secondary employer, having regard to the nature and extent of the secondary strikes.8 It was ultimately a proportionality assessment to determine whether the harm caused by the secondary strikes to the secondary employer was proportional to their impact on the business of the primary 6 Labour Court judgment above n 2 at para 16. 7 Id. 8 Id at para 17. 9 PILLAY AJ employer.9 If the harm was proportional, then the secondary strikes satisfied the requirement of section 66(2)(c); if it was not, then it was unprotected.10 [19] Based on SALGA II,11 the Labour Court noted that “it is an enquiry into the extent of the pressure that is placed on the primary employer”.12 This factual enquiry was to be undertaken with reference to each individual secondary employer; the grouping together of secondary employers in an industry to gauge for the combined effect of the secondary strikes on the primary employer was impermissible.13 A collective approach would deprive each secondary employer of the protection afforded to it by section 66(2)(c).14 [20] Furthermore, the enquiry into reasonableness required a consideration of the form and duration of the anticipated secondary strike including the extent of the strike’s impact, the sector in which it occurred, the number of employees involved and their conduct. In conducting a proportionality analysis to assess the reasonableness of the secondary strikes on the respondents, the Labour Court held that the secondary strikes would result in severe disruptions to the business of the secondary employers.15 It found AMCU’s suggestion that they use replacement labour impossible to implement. The lengthy recruitment process would far exceed the seven-day notice to secondary employers.16 Furthermore, AMCU’s apparent inability to control its members suggested that the secondary employers’ fears of violence during the secondary strikes were well founded.17 9 Id. 10 Id. 11 SALGA v SAMWU [2011] ZALAC 4; (2011) 32 ILJ 1886 (LAC) (SALGA II) at para 9. 12 Labour Court judgment above n 2 at para 18. 13 Id at para 22. 14 Id. 15 Id at para 198. 16 Id at para 199. 17 Id at para 207. 10 PILLAY AJ [21] On the law, the Labour Court acknowledged that the effect of the secondary strikes on the primary employer did not necessarily mean that there had to be a nexus between the primary and secondary employers. It held, however, that it was “difficult to conceive how a secondary strike could have a direct or indirect effect on the primary employer’s business without some relationship of sorts between the two employers”.18 On the facts, the Labour Court concluded that the anticipated secondary strikes would have no direct or indirect effect on Sibanye’s business.19 Accordingly, the Labour Court declared the secondary strikes unprotected with no order as to costs.20 Aggrieved by the outcome, AMCU applied for leave to appeal, which the Labour Court refused with costs. Labour Appeal Court [22] In refusing the petition for leave to appeal, the Labour Appeal Court relied on section 16(2)(a)(ii) of the Superior Courts Act,21 to find no exceptional circumstances and no significant point of law in the appeal since both parties had agreed that the matter had been rendered moot.22 It dismissed the appeal with costs, including the costs of two counsel.23 18 Id at para 219 with reference to SALGA v SAMWU [2008] 1 BLLR 66 (LC) (SALGA I). 19 Id at para 262. 20 Id at para 278. 21 10 of 2013. 22 Labour Appeal Court judgment above n 2 at para 24. 23 Id at para 26. 11 PILLAY AJ [23] However, the Labour Appeal Court endorsed the interpretation in both SALGA I24 and SALGA II25 that section 66(2)(c) imports a proportionality test to “weigh the effect of the secondary strike on the secondary employer and the effect of the nature and extent of the secondary strike on the business of the primary employer”.26 Referring to the judgment of the Labour Court, the Labour Appeal Court settled on the proportionality principle in the following terms: “There can be little doubt that in this paragraph the learned Judge clearly placed emphasis upon economic consequences of a secondary strike for the secondary employer. The weight to be accorded to this set of factors in the balancing exercise is a factual determination which will vary from case to case. The jurisprudence clearly mandates a proportionality assessment in which a court is enjoined to weigh the reasonableness, nature and extent of the secondary strike against the effect of the secondary strike on the business of the primary employer. This establishes that the economic consequences for the secondary employer must be taken into account. It follows that no further judgment of this Court is required to set out the nature of the proportionality enquiry. Manifestly, there may be different factual applications of this enquiry. The evaluation of the outcome thereof will depend upon a live dispute which would then require the attention of the Labour Court and, if necessary, this Court.”27 24 Id at para 20, citing SALGA 1 above n 18 at para 16: “In short, whether or not a secondary strike is protected is determined by weighing up two factors – the reasonableness of the nature and extent of the secondary strike (this is an enquiry into the effect of the strike on the secondary employer and will require consideration, inter alia, of the duration and form of the strike, the number of employees involved, their conduct, the magnitude of the strike’s impact on the secondary employer and the sector in which it occurs) and, secondly, the effect of the secondary strike on the business of the primary employer, which is, in essence, an enquiry into the extent of the pressure that is placed on the primary employer.” 25 Labour Appeal Court judgment id at para 21, citing SALGA 1I above n 11 at para 10 where it held that “[u]nder the head of proportionality, the court must weigh the effect of the secondary strike on the secondary employer and the effect of the nature and extent of the secondary strike on the business of the primary employer”. The remainder of para 10 of SALGA II provides: “The sub-section does not require actual harm to be suffered by the primary employer but that there must be the possibility that it may. The harm that the employer may suffer is not required to be direct. It may be harm that indirectly affects the business of the primary employer. It would, therefore, in every case require a factual inquiry to determine whether or not the possible effect the secondary strike will have on the business of the primary employer is reasonable. The harm that may be suffered by the secondary employer must be proportional to the possible effect the secondary strike may have on the business of the primary employer.” 26 Labour Appeal Court judgment id at para 21, quoting from SALGA II above n 11 at para 10. 27 Id at para 24. 12 PILLAY AJ In this Court Applicants’ submissions [24] AMCU submits that harm to the secondary employers is not a consideration under section 66(2)(c), properly interpreted. At least, it should not be given as much weight as the Labour Court accorded to it. Further, grouping together the secondary employers to assess the harm caused to the primary employer by the secondary strikes, is the incorrect approach and contrary to SALGA I. Additionally, AMCU argues that the Labour Court’s approach in finding that, based on past behaviour, AMCU members would likely resort to violence which amounted to collective guilt, is irreconcilable with the constitutional right to engage in strike action. AMCU relies on the judgments of the Labour Court in Hextex28 and Billiton,29 which disavowed a proportionality assessment. Respondents’ submissions [25] The second respondent objects to this Court’s jurisdiction being engaged. It contends that no constitutional issue is raised. Furthermore, no interpretation of section 66(2)(c) is necessary because, in its view, SALGA I sets out the correct approach to applying the section. It submits further that the matter is moot because the strike action has since ceased, and thus, the matter does not warrant consideration by this Court. [26] The third, fourth, sixth, seventh, eighth and tenth respondents accept that this matter raises constitutional issues and that this Court has jurisdiction. However, they contend that the appeal should have been lodged against the decision of the Labour Appeal Court, not that of the Labour Court. They resist the application on the further grounds of mootness, that it is not of general public importance, and that in the light of SALGA I and SALGA II, the applicants’ argument that section 66(2)(c) does not require harm nor does it import a proportionality analysis, lacks prospects of success. 28 Hextex v SA Clothing and Textile Workers Union (2002) 23 ILJ 2267 (LC). 29 Billiton Aluminium SA Ltd v National Union of Metalworkers of SA (2001) 22 ILJ 2434 (LC) (Billiton I). 13 PILLAY AJ Jurisdiction, mootness and leave to appeal [27] This dispute involves the interpretation of provisions of the LRA which give effect to section 23 of the Bill of Rights.30 It engages this Court’s jurisdiction. [28] Initially, AMCU brought its application for leave to appeal against the judgment of the Labour Court only. Now AMCU seeks leave to appeal against the judgments of both the Labour Court as well as the Labour Appeal Court. But AMCU does so only in its heads of argument. AMCU makes no formal application for condonation. Unsurprisingly, the respondents31 oppose the application for leave to appeal on the grounds of both non-compliance with the rules and mootness. [29] Although the application does not comply with rule 19(2) and (3) of the Rules of this Court, the respondents had notice in AMCU’s heads of argument of the latter’s intention to include the judgment of the Labour Appeal Court in this application for leave to appeal. Furthermore, full argument was presented on the discrete question of law. [30] The Labour Appeal Court concluded that the matter was moot.32 I agree. However, the interests of justice standard applies for determining both condonation and whether a moot matter should be heard.33 If leave to appeal is granted, the decision would clarify the approach to secondary strikes, for the parties and others in the labour relations community. A judgment would also resolve disputes, if any, between different 30 National Education Health and Allied Workers Union v University of Cape Town [2002] ZACC 27; 2003 (3) SA 1 (CC); 2003 (2) BCLR 154 (CC) at para 14. See also section 167(3)(b)(i) of the Constitution. 31 The second, third, fourth, sixth, seventh, eighth and tenth respondents. 32 Trencon Construction (Pty) Limited v Industrial Development Corporation of South Africa Limited [2015] ZACC 22; 2015 (5) SA 245 (CC); 2015 (10) BCLR 1199 (CC) at para 83. 33 Normandien Farms (Pty) Limited v South African Agency for Promotion of Petroleum Exportation and Exploitation SOC Limited [2020] ZACC 5; 2020 (4) SA 409 (CC); 2020 (6) BCLR 748 (CC) at para 48 and President of the Republic of South Africa v Democratic Alliance [2019] ZACC 35; 2020 (1) SA 428 (CC); 2019 (11) BCLR 1403 (CC) at para 17. 14 PILLAY AJ courts.34 The issue is both important and complex. The prospects of success are reasonable.35 Leave to appeal is therefore granted against the judgments of both the Labour Court and the Labour Appeal Court. The issues [31] Essentially, this appeal turns on the substantive requirements for lawful secondary strikes in terms of section 66(2)(c).36 Specifically, the question raised is whether section 66(2)(c) imports the principle of proportionality in assessing the lawfulness of secondary strikes. More specifically, the question is whether section 66(2)(c) factors in a balancing of the impact of secondary strikes on secondary employers, on the one hand, with their effect on the primary employer on the other hand. Thus, if secondary strikes impact disproportionately harshly on secondary employers, as a party uninvolved in the primary strike, would secondary employers be entitled to interdict the secondary strikes under section 66(3)? [32] AMCU raises two subsidiary issues relating to the form of secondary strikes. First, in assessing the reasonableness of multi-employer secondary strikes, should the impact on secondary employers be considered individually or collectively? Second, should interdicts against violence in the past and violence during the primary strike be taken as factors informing the assessment of compliance with section 66(2)(c)? [33] As the matter is moot, the brief summary of the facts and arguments above are intended to set the scene for what is exclusively an exercise in interpreting sections 66(2)(c) and (3). Any references to the facts will be purely to contextualise 34 MEC for Education, Kwazulu-Natal v Pillay [2007] ZACC 21; 2008 (1) SA 474 (CC); 2008 (2) BCLR 99 (CC) (Pillay) at para 32 and Independent Electoral Commission v Langeberg Municipality [2001] ZACC 23; 2001 (3) SA 925 (CC); 2001 (9) BCLR 883 (CC) at para 11. 35 POPCRU v SACOSWU [2018] ZACC 24; 2019 (1) SA 73 (CC); 2018 (11) BCLR 1411 (CC) at para 44. 36 Section 66(1) of the LRA reads: “In this section ‘secondary strike’ means a strike, or conduct in contemplation or furtherance of a strike, that is in support of a strike by other employees against their employer, but does not include a strike in pursuit of a demand that has been referred to a council if the striking employees, employed within the registered scope of that council, have a material interest in that demand.” 15 legal principles. For this interpretative exercise, my starting point is the Constitution, which in turn invites me to look to international law. Together, they underpin the PILLAY AJ interpretation. Section 23 of the Constitution [34] The analysis of section 66(2)(c) begins with an appreciation that the LRA seeks to give effect to section 23 of the Constitution.37 Contextualising the interpretation within the constitutional framework is necessary to acknowledge the elevated status that collective bargaining and the right to strike enjoy under our current democratic dispensation. The right to strike is constitutionalised because it is a recognised human rights response to slavery, servitude and forced labour. 37 Section 23 of the Constitution reads: “(1) (2) (3) (4) (5) (6) to form and join an employers’ organisation; and to participate in the activities and programmes of an employers’ organisation. Everyone has the right to fair labour practices. Every worker has the right (a) to form and join a trade union; (b) to participate in the activities and programmes of a trade union; and (c) to strike. Every employer has the right (a) (b) Every trade union and every employers’ organisation has the right (a) (b) (c) Every trade union, employers' organisation and employer has the right to engage in collective bargaining. National legislation may be enacted to regulate collective bargaining. To the extent that the legislation may limit a right in this Chapter, the limitation must comply with section 36(1). National legislation may recognise union security arrangements contained in collective agreements. To the extent that the legislation may limit a right in this Chapter, the limitation must comply with section 36(1).” to determine its own administration, programmes and activities; to organise; and to form and join a federation. 16 PILLAY AJ [35] Self-evidently, the preamble to the LRA, the LRA’s purpose outlined in section 1,38 and its approach to interpretation articulated in section 3,39 acknowledge the supremacy of the Constitution. These provisions, read with chapters two, three, four and six of the LRA respectively, regulate freedom of association, the form and functioning of collective bargaining, the regulation of strikes and lockouts and the regulation of trade unions and employers’ organisations. Together, they establish the fundamentals to enable effective collective bargaining foreshadowed in section 23(5)-(6) of the Constitution. [36] Conceptually, the scheme of the LRA does not create a statutory duty to compel bargaining at particular levels (e.g., at plant, company or national levels). Nor does it prescribe what issues should be bargained. Instead, the model adopted “allows the parties, through the exercise of power, to determine their own arrangements. The exercise of power, or indeed persuasion, is given statutory impetus by the draft Bill’s provision for organisational rights and a protected right to strike.”40 In other words, the 38 Section 1 of the LRA states: (a) (b) “The purpose of this Act is to advance economic development, social justice, labour peace and the democratisation of the workplace by fulfilling the primary objects of this Act, which are— to give effect to and regulate the fundamental rights conferred by section 27 of the Constitution; to give effect to obligations incurred by the Republic as a member state of the International Labour Organisation; to provide a framework within which employees and their trade unions, employers and employers’ organisations can— (i) collectively bargain to determine wages, terms and conditions of employment and other matters of mutual interest; and formulate industrial policy; and (c) (d) (ii) to promote— (i) (ii) (iii) (iv) orderly collective bargaining; collective bargaining at sectoral level; employee participation in decision-making in the workplace; and the effective resolution of labour disputes.” 39 Section 3 of the LRA states: “Any person applying this Act must interpret its provisions— (a) (b) (c) to give effect to its primary objects; in compliance with the Constitution; and in compliance with the public international law obligations of the Republic.” 40 Explanatory Memorandum to the Draft Negotiating Document in the Form of a Labour Relations Bill GN 97 GG 16259, 10 February 1995 (Explanatory Memorandum) at 121. Below, I explain my reliance on the Explanatory Memorandum. 17 PILLAY AJ scheme renders the right to bargain collectively in section 23 organisationally enforceable. [37] This model of collective bargaining in the LRA seeks to create a particular balance to level the rights of trade unions with employers to enable the effective exercise of power during collective bargaining. Collective bargaining is pre-eminently the means for determining the terms and conditions of employment. Beyond a basic floor of statutory rights, wages and working conditions are determined largely by collective bargaining. Thus, sight should not be lost of the instrumental and transformative role of collective bargaining processes to remedy inequality, discrimination and poverty in the workplace. Consequently, courts must be cautious when interpreting, applying and limiting the constitutional rights to bargain collectively and to strike, as enabled under the LRA. Impeding employees’ rights to bargain collectively and to strike, could blunt their principal weapons in the struggle to improve their livelihood to overcome centuries of discrimination. Economic hardship for the employer is a predictable consequence of strikes. So is the standard “no work, no pay” rule for the employees. Notwithstanding this, the LRA institutionalises power-play as a driver of collective bargaining.41 Therefore, prudence must prevail before the rights to bargain collectively and to strike are weakened. [38] That said, no right, including the right to strike, is absolute. Limitations on the right to participate in a secondary strike arise from both the Constitution and the LRA, although any limitation must comply with section 36 of the Constitution. And when interpreting the LRA, the Court must promote the spirit, purport and objects of the Bill of Rights.42 Additionally, three provisions direct us to seek out the stance of the International Labour Organisation (ILO) on secondary strikes. 41 Ex parte Chairman of the Constitutional Assembly: In re Certification of the Constitution of the Republic of South Africa, 1996 [1996] ZACC 26; 1996 (4) SA 744 (CC); 1996 (10) BCLR 1253 (CC) at 795. 42 Section 39(2) of the Constitution. 18 PILLAY AJ International law [39] Section 233 of the Constitution states: “When interpreting any legislation, every court must prefer any reasonable interpretation of the legislation that is consistent with international law over any alternative interpretation that is inconsistent with international law.” [40] Section 1(b) of the LRA provides: “The purpose of this Act is to advance economic development, social justice, labour peace and the democratisation of the workplace by fulfilling the primary objects of this Act which are— . . . (b) to give effect to obligations incurred by the Republic as a member state of the International Labor Organisation.” [41] Section 3 of the LRA directs that— “(a)ny person applying this Act must interpret its provisions— (a) (b) (c) to give effect to its primary objects; in compliance with the Constitution; and in compliance with the public international law obligations of the Republic.” [42] On 19 February 1996, South Africa ratified the ILO Conventions on Freedom of Association and Protection of the Right to Organise (Convention No. 87)43 and the Right to Organise and Collective Bargaining (Convention No. 98).44 Both conventions remain in force and are binding. 43 Freedom of Association and Protection of the Right to Organise Convention, 9 July 1948. 44 Right to Organise and Collective Bargaining Convention, 1 July 1949. 19 PILLAY AJ [43] Neither convention expressly confers a right to strike. However, articles 3 and 10 of Convention No. 8745 inspire both the Committee on Freedom of Association and the Committee of Experts on the Application of Conventions and Recommendations (“the Committee of Experts”) to progressively develop principles on the right to strike. Article 3 sets out the right of workers’ organisations to organise their activities and to formulate their programmes. Article 10 recognises the objective of workers’ organisations to further, and defend, the interests of workers.46 The Committee of Experts acknowledges the competing interests implicated during strikes when it observes: “Strikes are essential means available to workers and their organisations to protect their interests, but there is a variety of opinions in relation to the right to strike. While it is true that strike action is a basic right, it is not an end in itself, but the last resort for workers’ organisations, as its consequences are serious, not only for employers, but also for workers, their families and organisations and in some circumstances for third parties.”47 [44] Concerning secondary strikes, the Committee of Experts adopts the following stance: “With regard to so-called ‘sympathy’ strikes, the Committee considers that a general prohibition of this form of strike action could lead to abuse, particularly in the context of globalisation characterised by increasing interdependence and the internationalisation of production, and that workers should be able to take such action, provided that the initial strike they are supporting is itself lawful.”48 45 International Labour Organisation “Giving Globalisation a Human Face” (conference report presented at the 101st session of the International Labour Conference, Geneva, 2012) at 117 and International Labour Organisation “Freedom of Association: Digest of Decisions and Principles of the Freedom of Association Committee of the Governing Body of the ILO” 5 ed (International Labour Office, Geneva 2006) (Freedom of Association: Digest of Decisions) at paras 520-7. 46 Giving Globalisation a Human Face id at 117. 47 Id (emphasis added). 48 Id at para 125. 20 [45] Similarly, the Committee on Freedom of Association regards secondary strikes to be protected by international labour law. In its opinion— PILLAY AJ “[a] general prohibition of sympathy strikes could lead to abuse and workers should be able to take such action provided the initial strike they are supporting is itself lawful.”49 [46] By 2017, the ILO position was categorical: “The position of the ILO on this issue has been very clear: sympathy strikes should be permitted subject to the sole requisite that the original strike is lawful.”50 [47] This is not to suggest that sympathy strikes should be afforded utterly unfettered protection. The Committee of Experts recognises the importance of sympathy strikes but emphasises that the justification for recourse to this type of strike should be specified.51 Typical of international standards, the ILO recommends a minimum requirement for sympathy strikes. [48] At this point, it is convenient to note the following explanation that Cooper tenders for the different terminology: “In broad terms, what is alternatively called sympathy, secondary or solidarity action refers to industrial action taken by workers of an employer who is not directly involved in a dispute (the secondary employer), in support of workers employed by the employer in the dispute (the primary employer). The terms secondary, sympathy or solidarity, although often used interchangeably, are not ideologically neutral: the terms ‘sympathy’ and ‘solidarity’ tend to emphasise workers’ interests, while the term ‘secondary’ emphasises the relationship between employers.”52 49 Freedom of Association: Digest of Decisions above n 45 at para 534. 50 García “The Right to Strike as a Fundamental Human Right: Recognition and Limitations in International Law” (2017) 44 Revista Chilena de Derecho 781 at 796-7. 51 Cooper “Sympathy Strikes” (1995) 16 Industrial Law Journal 759 at 759, citing International Labour Organisation “Report of the Committee of Experts Freedom of Association and Collective Bargaining” (conference report presented at the 81st session of the International Labour Conference, Geneva 1994) at 74. 52 Cooper id at fn 1. 21 PILLAY AJ [49] The Committee of Experts, established in 1926, monitors compliance by member states with the conventions and recommendations. It offers impartial, technical assessments regarding the status of the application of international labour standards.53 As such, its opinions are advisory or “soft law”.54 Of course, underestimating the considerable moral authority of these supervisory bodies in the international arena would be a mistake.55 Nevertheless, the Committee of Experts acknowledges that since 1990 its terms of reference do not allow it to give “definitive interpretations of Conventions, competence to do so being vested in the International Court of Justice by article 37 of the Constitution of the ILO”.56 Its “opinions and recommendations are not binding within the ILO supervisory process and are not binding outside the ILO unless an international instrument expressly establishes them as such or the Supreme Court of a country so decides of its own volition”.57 The stance of the Committee of Experts is that “in so far as its views are not contradicted by the International Court of Justice, they should be considered as valid and generally recognised”.58 [50] Enforcement of ILO standards is achieved by member states, including our own, ratifying the standards and giving effect to them in national constitutions: “Freedom of association is guaranteed in almost all national Constitutions. In most cases, the Constitution provides that this freedom shall be exercised in accordance with the law. Most Constitutions also provide for a number of exceptions, in particular in relation to security forces, or leave the determination of exceptions to the law. Specific references to the right to freely form, to join or not to join trade unions are also found 53 International Labour Organisation “Application of International Labour Standards 2020: Report of the Committee of Experts on the Application of Conventions and Recommendations” (conference report presented at the 109th session of the International Labour Conference, Geneva, 2020) (Application of International Labour Standards) at 11. 54 International Labour Organisation “Collective Bargaining in the Public Service: A Way Forward” (conference report presented at the 102nd session of the International Labour Conference, Geneva, 2013) at 2. 55 Application of International Labour Standards 2020 above n 53 at 11. 56 International Labour Organisation “Report of the Committee of Experts on the Application of Conventions and Recommendations” (conference report presented at the 102nd session of the International Labour Conference, Geneva 2013) at 10. 57 International Labour Organisation “Collective Bargaining in the Public Service: A Way Forward” (conference report presented at the 102nd session of the International Labour Conference, Geneva 2013) at 2. 58 Id at 2. 22 in 142 Constitutions. Furthermore, the right to strike is constitutionally protected in 93 countries.”59 PILLAY AJ [51] Significantly, fewer countries constitutionalise the right to strike than the right to freedom of association. And those countries that permit secondary strikes regulate them to a greater degree than primary strikes.60 Comparative information about secondary strikes “reveals a spectrum of national positions, ranging from a broadly permissive stance in countries such as Greece, Finland, Norway and Sweden, to those that do not recognise or permit it”, such as the United Kingdom.61 In those countries that do permit secondary strikes, something more than the bare minimum of the lawfulness of the original strike is anticipated. Conceptually, secondary strikes distinguish themselves by virtue of their relationships with the primary strike. In countries like Spain, Italy and France, which constitutionalise the right to strike, the requirement ranges “from a professional or occupational interest” to “a sufficient interest”.62 Significantly, “[i]n all three jurisdictions the courts have played a key role in giving content to the definition, at times liberalising and at other times limiting it”.63 [52] Before considering how South Africa gives effect to the constitutional right to strike in the LRA, a slight detour to the United Kingdom will give some perspective to the degree to which South Africa complies with ILO standards and the opinions of the Committee of Experts on secondary strikes. [53] Since 1989, the Committee of Experts repeatedly criticised the United Kingdom’s non-observance of Convention No. 87 by outlawing secondary 59 Giving Globalisation a Human Face above n 45 at 4 and 50, specifically fn 264. 60 Seady and Thompson “Strikes and Lockouts” in Thompson and Benjamin South African Labour Law (2020) 1 at 301. 61 National Union of Rail, Maritime and Transport Workers v the United Kingdom [2014] ECHR 366 (RMT v UK) at para 91. 62 Cooper above n 51 at 764. 63 Id. 23 PILLAY AJ strikes.64 The European Committee of Social Rights has also criticised this stance.65 Matters came to a head in RMT v UK. The trade union, RMT, with a membership of more than 80 000 employees in the transport industry, challenged the blanket ban on secondary strikes.66 [54] On the principle of proportionality, RMT contended that the absolute ban completely eschewed any balancing of the competing rights and interests, countenancing no differentiation between situations. In contrast, the Government defended the Legislature’s preference for a uniform rule – a total ban – persisting that a less restrictive rule would be impracticable and ineffective.67 [55] The European Court of Human Rights (ECtHR) held that the ban on secondary strikes in the United Kingdom was an interference with the right to freedom of association in article 11 of the European Convention on Human Rights68 (ECHR), thus acknowledging indirectly that the right to strike, as embodied in the right to freedom of association, includes secondary strikes. However, the ECtHR went on to qualify that the interference was justified. It found that interference with freedom of association in the circumstances of that case was not especially far-reaching. Having regard to “the breadth of the margin of appreciation in this area” the Court rejected RMT’s arguments 64 RMT v UK above n 61 at paras 32-3. 65 Id at paras 35-7. 66 Id at paras 6-7. 67 Id at para 100. 68 Freedom of association under article 11 of the European Convention on Human Rights, 4 November 1950 (ECHR) reads as follows: “(1) (2) Everyone has the right to freedom of peaceful assembly and to freedom of association with others, including the right to form and to join trade unions for the protection of his interests. No restrictions shall be placed on the exercise of these rights other than such as are prescribed by law and are necessary in a democratic society in the interests of national security or public safety, for the prevention of disorder or crime, for the protection of health or morals or for the protection of the rights and freedoms of others. This article shall not prevent the imposition of lawful restrictions on the exercise of these rights by members of the armed forces, of the police or of the administration of the State.” 24 on trade union solidarity and efficacy.69 Preferring the Government’s position, the ECtHR reasoned that— PILLAY AJ “by its nature secondary action may well have much broader ramifications than primary action. It has the potential to impinge upon the rights of persons not party to the industrial dispute, to cause broad disruption within the economy and to affect the delivery of services to the public.”70 [56] The ECtHR concluded that the absolute ban on solidarity strikes was not disproportionate, but was necessary in a democratic society. In passing, the margin of appreciation doctrine is at best doubtful, if not irrelevant for our purposes, considering that our approaches to constitutional limitation and interpretation are prescribed under sections 36 and 39 of the Constitution. Furthermore, it will become clear, if it is not already so from the exposition of the purpose and aims of the LRA above, that Parliament was at pains to give effect to the Constitution and international law. Interpreting the LRA [57] In Mansingh,71 this Court helpfully recalled the approach to interpretation of our negotiated Constitution: “It is well-established that courts need not look to the drafter’s intention when engaging in constitutional (or statutory) interpretation. However, as stated above, we must adopt a purposive reading of section 84(2)(k). When there is documentary evidence regarding that purpose, we may, in appropriate circumstances, have regard to such 69 The doctrine of the margin of appreciation has its genesis in international human rights law, and has evolved through the jurisprudence of the ECtHR. This legal doctrine is used to assess whether state parties to the ECHR should be sanctioned for limiting the enjoyment of rights. The assessment involves a balancing of individual and group interests with the national interest to resolve conflicts. Underpinning the margin of appreciation doctrine are principles of deference, subsidiarity and precedent. Insofar as the doctrine can impede constitutional transformation, it should be rejected. See, in respect of the doctrine generally and its specific applications Kratochvíl “The Inflation of the Margin of Appreciation by the European Court of Human Rights” (2011) 29 Netherlands Quarterly of Human Rights 324. 70 RMT v UK above n 61 at para 82. 71 Mansingh v General Council of the Bar [2013] ZACC 40; 2014 (2) SA 26 (CC); 2014 (1) BCLR 85 (CC) citing Natal Joint Municipal Pension Fund v Endumeni Municipality [2012] ZASCA 13; 2012 (4) SA 593 (SCA) at paras 17-26. 25 PILLAY AJ evidence – the travaux préparatoires. To the extent that the intention of the panel of experts is relevant, it supports the reasoning set out above.”72 [58] Recourse to travaux préparatoires (preparatory works)73 was first had in Makwanyane74 in relation to the Constitution: “Our Constitution was the product of negotiations conducted at the Multi-Party Negotiating Process. The final draft adopted by the forum of the Multi-Party Negotiating Process was, with few changes, adopted by Parliament. The Multi-Party Negotiating Process was advised by technical committees, and the reports of these committees on the drafts are the equivalent of the travaux préparatoires relied upon by the international tribunals. Such background material can provide a context for the interpretation of the Constitution and, where it serves that purpose, I can see no reason why such evidence should be excluded. The precise nature of the evidence, and the purpose for which it may be tendered, will determine the weight to be given to it.”75 [59] This Court continues to support appropriate recourse to preparatory works: “Background evidence may, however, be useful to show why particular provisions were or were not included in the Constitution. It is neither necessary nor desirable at this stage in the development of our constitutional law to express any opinion on whether it might also be relevant for other purposes, nor to attempt to lay down general principles governing the admissibility of such evidence. It is sufficient to say that where the background material is clear, is not in dispute, and is relevant to showing why particular provisions were or were not included in the Constitution, it can be taken into account by a Court in interpreting the Constitution. These conditions are satisfied in the present case.”76 72 Mansingh id at para 27. 73 In Mansingh id at fn 50, this Court defined the “travaux préparatoires” as “the official documents recording the negotiations, drafting and discussions during the process of creating a legal instrument or Constitution”. 74 S v Makwanyane [1995] ZACC 3; 1995 (3) SA 391 (CC); 1995 (6) BCLR 665 (CC) (Makwanyane). 75 Id at para 17 (emphasis added). 76 Mansingh above n 71 at para 27 quoting Makwanyane id at para 19 (emphasis added). 26 PILLAY AJ [60] My purpose for having recourse to the Explanatory Memorandum is limited to acknowledging the genesis of the LRA as a “[d]raft negotiating document in the form of a Labour Relations Bill”,77 to glean insights into the meaning of section 66(2)(c). Cooper78 explains: “Arising from concerns by the Minister of Labour and employers that solidarity action needs to be clearly defined in the law, proposals were made in NEDLAC which significantly modified the draft Bill’s provisions regarding sympathy strikes. Underlying the proposals is the compromise agreement between labour and business that a balance should be struck between the functionality of sympathy strikes and their meaningfulness. In other words while they should be functional to collective bargaining and thus restricted in some way, this should not undermine their purpose as instruments of worker solidarity. The proposed provisions sought to limit solidarity strikes through requiring that there be an economic connection between the primary and secondary employers for a protected solidarity strike. In other words what was envisaged was the introduction of the ally doctrine.”79 [61] In the LRA the negotiating partners jettisoned the ally doctrine because it severely restricted secondary strikes to the point of virtually outlawing them.80 Narrowly defined, an ally would be an associated or take-over entity of the primary employer or an entity having a direct link with the primary employer. Somewhere between the outright outlawing of secondary strikes and altruism, the negotiating partners sought to attain a balance. Unsurprisingly, the LRA shows traces of the contestation for the best possible dispensation by the negotiating partners for their respective constituencies. Section 66(2)(c) does not escape this general observation. 77 Stated in the full title of the Labour Relations Bill, namely “Draft Negotiating Document in the Form of a Labour Relations Bill”. 78 The Explanatory Memorandum above n 40 at 112 records Ms Cooper to be a researcher who assisted the Ministerial Legal Task Team that was formed “to overhaul the laws regulating labour relations and to prepare a negotiating document in draft Bill form to initiate a process of public discussion and negotiation by organised labour and business and other interested parties”. 79 Cooper above n 51 at 782. 80 Id at 762-3. 27 PILLAY AJ Thus, when interpreting the LRA, fidelity to both the text81 and the context of the negotiation is necessary. [62] In his lucid dissent (the second judgment), my colleague Jafta J’s literal interpretation of section 66(2)(c) accords closely with the ILO’s position in saying that secondary strikes should be subject only to the requirement that the primary strike be lawful.82 However, a literal interpretation misses out on the nuances to the text that have their genesis in the context of a “Draft Negotiating Document in the Form of a Labour Relations Bill” and a constitutionalised right to strike. Furthermore, the ILO standard sets a minimum requirement. Manifestly, the ILO prescribes neither the procedural prerequisites for embarking on secondary strikes or any other substantive requirements. Those are matters for sovereign states to determine in national legislation. Section 66(2)(c) meets the ILO’s minimum requirement that the primary strike must be lawful but adds procedural and other requirements consistent with the negotiations and the Constitution. Strikes under the LRA [63] The scheme of the LRA, in regulating strikes, aims to provide clear procedural requirements. Once they are met, employees may lawfully embark on peaceful strikes. Interference in the power-play between parties engaged in a lawful strike is off limits and limiting collective bargaining that complies with the rules of engagement is a hard row to hoe under the LRA. [64] Secondary strikes replicate the pre-eminence of collective bargaining and the right to strike. Once a trade union fulfils the formal requirements prescribed in section 66(2)(a) and (b), paragraph (c) stipulates the substantive requirements. Unlike 81 Bertie van Zyl (Pty) Ltd v Minister for Safety and Security [2009] ZACC 11; 2010 (2) SA 181 (CC); 2009 (10) BCLR 978 (CC) (Bertie van Zyl) at para 22. 82 See [44]. 28 PILLAY AJ the broad scope of common law interdicts,83 the text of section 66(3) narrows the grounds for interdicts as follows: “Subject to section 68(2) and (3), a secondary employer may apply to the Labour Court for an interdict to prohibit or limit a secondary strike that contravenes subsection (2).” [65] Section 68(2) and (3)84 regulate notice and other procedural requirements for interdicting a strike. And interdicting a secondary strike that complies with the requirements of section 66(2)(c) is impermissible. [66] When assessing compliance with section 66(2)(c), one cannot afford to overlook the differences between strikes85 and secondary strikes.86 Substantively, a key 83 Hotz v University of Cape Town [2016] ZASCA 159; 2017 (2) SA 485 (SCA) at para 29 states: “The law in regard to the grant of a final interdict is settled. An applicant for such an order must show a clear right; an injury actually committed or reasonably apprehended; and the absence of similar protection by any other ordinary remedy. Once the applicant has established the three requisite elements for the grant of an interdict the scope, if any, for refusing relief is limited. There is no general discretion to refuse relief.” (Emphasis added.) 84 Section 68(2)-(3) of the LRA reads: “(2) The Labour Court may not grant any order in terms of subsection (1)(a) unless 48 hours’ notice of the application has been given to the respondent: However, the Court may permit a shorter period of notice if— (a) (b) (c) the applicant has given written notice to the respondent of the applicant’s intention to apply for the granting of an order; the respondent has been given a reasonable opportunity to be heard before a decision concerning that application is taken; and the applicant has shown good cause why a period shorter than 48 hours should be permitted. (3) Despite subsection (2), if written notice of the commencement of the proposed strike or lock-out was given to the applicant at least 10 days before the commencement of the proposed strike or lock-out, the applicant must give at least five days’ notice to the respondent of an application for an order in terms of subsection (1)(a).” 85 Section 213 of the LRA defines “strike” to mean: “[T]he partial or complete concerted refusal to work, or the retardation or obstruction of work, by persons who are or have been employed by the same employer or by different employers, for the purpose of remedying a grievance or resolving a dispute in respect of any matter of mutual interest between employer and employee, and every reference to ‘work’ in this definition includes overtime work, whether it is voluntary or compulsory.” 86 Section 66(1) of the LRA defines “secondary strike” to mean: “[A] strike, or conduct in contemplation or furtherance of a strike, that is in support of a strike by other employees against their employer, but does not include a strike in pursuit of a demand that has been referred to a council if the striking employees, employed within the registered scope of that council, have a material interest in that demand.” 29 PILLAY AJ distinction is that employees engaged in secondary strikes cannot bargain with their own employer for their own benefits. By definition, the secondary strike must be in support of the primary strike. While the secondary strike itself is intended to induce the primary employer to bargain with its employees, nothing stops the secondary employer and its employees engaging each other to achieve that inducement. In other words, they do not have to be passive bystanders, helplessly weathering the secondary strike. [67] The procedural requirements for a strike are infinitely more onerous than for a secondary strike. A strike must be preceded by conciliation, deadlock and notice to the employer. All this takes time before a strike can commence. Four months lapsed between AMCU commencing negotiations and the primary strike. In the case of secondary strikes, no conciliation, no deadlock, and no dialogue are prescribed as prerequisites. Once the primary strike is lawful, all that remains to be done is for the trade union to give seven days’ notice to the secondary employer of its employees’ intention to embark on a secondary strike.87 Cumulatively, the absence of prior engagement, the brevity of the notice and the fact that the secondary employer and its employees have no interest in the outcome of the primary strike, distinguish the secondary employer from the primary employer. Secondary employers having employees who belong to other trade unions that are not engaged in the secondary strike, adds another dynamic to a situation already complicated by multi-dimensional power-play. [68] Secondary employers, who have no direct or indirect effect on collective bargaining at the primary employer, have no control whatsoever over the process. Without section 66(3), the scheme of the LRA regarding secondary strikes would leave the secondary employer with little leverage to safeguard its interests. Section 66(3) is a shield in the hands of secondary employers who would otherwise have no means of protecting their enterprises. What it is not, is a weapon to annihilate the constitutional right to strike. Thus, when invoking section 66(3), ways of limiting the secondary strike 87 Section 66(2)(b) of the LRA. 30 PILLAY AJ should be investigated first before its outright prohibition is sought. In Samancor,88 the Labour Court gave its ruling on what would be proportional before anticipating a settlement agreement limiting the secondary strikes, which it made an order of that Court. [69] Section 66(2)(c) seeks to attain a balance between different models of regulating secondary strikes recognised by the ILO. To section 66(2)(c) I turn. Section 66(2)(c) of the LRA [70] SALGA I and SALGA II are currently authoritative decisions which maintain that section 66(2)(c) invites a proportionality assessment when considering whether a secondary strike can be said to be reasonable. Proportionality implies that a secondary strike that impacts more harshly on the secondary employer than on the primary employer would be disproportionate, and therefore, unreasonable and unlawful. Furthermore, if the harm to the secondary employer is so immense, then that fact alone should spare the secondary employer from enduring a secondary strike, even if the secondary strike has an effect on the primary employer. This case seeks to confirm or refute that interpretation of section 66(2)(c), which the Labour Court upheld. [71] Section 66(2)(a)-(c) reads: “(2) No person may take part in a secondary strike unless— (a) the strike to be supported complies with the provisions of sections 64 and 65; (b) the employer of the employees taking part in the secondary strike or, where appropriate, the employers’ organisation of which that employer is a member, has received written notice of the proposed secondary strike at least seven days prior to its commencement; and 88 Samancor Ltd v National Union of Metalworkers of SA (1999) 20 ILJ 2941 (LC) (Samancor) at paras 32-3 and 35. 31 PILLAY AJ (c) the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer.” [72] Unlike RMT v UK, in which the challenge was to the compliance of national legislation with international law,89 this application is not a challenge to the validity of section 66(2)(c), but rather concerns its interpretation and application. Section 66(2)(c) is intended to strike a balance between employee and employer rights, and between South Africa’s international and constitutional obligations. And the question is whether the Labour Court’s interpretation and application of section 66(2)(c) achieves the appropriate balance. [73] Section 66(2)(a) prescribes the procedural prerequisite that the primary strike must be lawful. This complies with the Committee of Experts’ stipulation of “the sole requisite that the original strike is lawful”.90 Typical of international law, that is a minimum standard to maximise assent from member states. The lawfulness of the primary strike is not the only requirement for a lawful secondary strike. Member states are free to regulate secondary strikes according to their national laws. Research has not revealed a single member state that imposes no other limitations for a secondary strike other than the lawfulness of the primary strike. [74] Embedded in section 66(2)(c) are at least six substantive requirements that regulate secondary strikes. These requirements operate as internal or built-in limitations on the right to participate in a secondary strike.91 [75] First, the critical requirement is for a secondary strike to have “an effect”. Having an effect on the primary employer is the baseline threshold requirement for a secondary strike. A secondary strike that can have no effect, that is entirely altruistic 89 Section 224 of the Trade Union and Labour Relations (Consolidation) Act 1992. 90 García above n 50 at 496-7. 91 Samancor above n 88 at para 2. 32 PILLAY AJ and amounts to “mindless exercises of worker solidarity for the sake of nothing but worker solidarity”,92 is impermissible. Such solidarity strikes fall on the opposite end of absolute bans on the secondary strike spectrum and are not a proportional exercise of the right to strike. Nor is an absolute ban a reasonable and justifiable limitation on the right to strike. Neither represent any balance whatsoever of the right to strike. Somewhere between these extremes section 66(2)(c) must find its place. [76] Second, the words “possible” and “indirect”, used in section 66(2)(c) in relation to “effect”, set the bar low and wide for a lawful secondary strike. The word “possible” lowers the threshold to mean something less than probable – not actual93 – but more than notional. What is required is that the secondary strike must be capable of having an effect. It is not necessary to predict with certainty that it necessarily would, but merely that it could, have an influence on the business of the primary employer.94 The word “indirect” expands the scope of the possible effect, thus shifting the balance to favour employee solidarity. However, that is about as far as the section inclines towards the solidarity side of the spectrum. The remaining words and phrases rebalance section 66(2)(c). [77] Third, the effect of the secondary strike must be “on the business” of the primary employer. Affecting the business of the primary employer anticipates a nexus between the business of the primary employer and the secondary strike, and by 92 SALGA I above n 18 at para 10 citing Grogan Collective Labour Law 1 ed (Juta & Co Ltd, Cape Town 2007). 93 SALGA I id at para 22 and SALGA II above n 11 at para 16. 94 Samancor above n 88 at para 24. See also SALGA II id at paras 16-7: “Mr Brassey also submitted that the secondary strike must bring its influence to bear in some tangible or material way on the secondary employer who must then put pressure on the primary employer to compromise or capitulate to the demands of its workers. I cannot agree. There is no requirement in section 66 of the Act that the secondary employer should exert influence on the primary employer or that the secondary employer should have the capacity to exert influence on the primary employer in order to encourage it to compromise or capitulate to the demands of the workers. What section 66 requires is that the secondary strike should have a possible direct or indirect effect on the business of the primary employer and that the nature and extent of the secondary strike should be reasonable in relation to the possible direct or indirect effect on the business of the primary employer. In conclusion the contention by the appellant that the secondary strike was unreasonable falls to be rejected.” 33 PILLAY AJ extension, the secondary employer and its employees. Impliedly, the nexus must be such that the secondary strike is reasonably capable of influencing the business of the primary employer. Such influence would be present, for instance, if there pre-exists a substantial commercial connection between both businesses. A supplier-customer relationship would establish a connection if the level of mutual interdependence is significant. This principle applied in Sealy,95 Samancor, Billiton,96 and Clidet.97 In SALGA I, the Labour Court found that a secondary strike at municipalities could have an effect on government at higher levels.98 Instructive is the Labour Appeal Court’s approach to unravelling the interconnectedness of enterprises in Samancor99 and 95 Sealy of South Africa (Pty) Ltd v Paper Printing Wood and Allied Workers Union (1997) 18 ILJ 392 (LC). 96 Billiton I above n 29 and Billiton Aluminium SA Ltd v National Union of Metalworkers of SA [2003] ZALAC 17; (2003) 24 ILJ 2259 (LAC) (Billiton II). 97 Clidet N.O. 957 (Pty) Ltd v SA Municipal Workers Union (2011) 32 ILJ 1070 (LC) at para 11 states as follows: “In these circumstances, I fail to appreciate what pressure will be placed on the business of MTC should the secondary action proceed. The only significant effect that the strike will have is the inconvenience to the thousands of commuters who rely on the applicant for their daily transport. They will be inconvenienced no doubt by the primary strike given the absence of cashiers and the like, but the effect of the secondary strike will be to deny them access to the transport on which they ordinarily rely. But the question here is not the extent of any inconvenience to commuters rather than whether on the test established by section 66, the applicant can be said to be reasonably capable of exerting pressure on MTC to meet the union’s demand that its employees should be permanently employed. For the above reasons, in my view, that question must be answered in the negative.” 98 See SALGA I above n 18 at para 21: “Given the integrated, co-ordinated and co-operative structure of government as a whole, it is entirely possible that the withdrawal of municipal services will have at least an indirect, if not a direct effect on the business of those higher levels of government engaged in the primary strike, and will thus place pressure on them in the national bargaining process currently underway.” 99 Samancor above n 88 at paras 29-31 states: “The two applicants appear to do business together. It appears that some of their plants are conducted jointly or fall within the scope of management of the first applicant, Samancor. However, Manganese Metals does not seem to play any significant role in the provisions of chrome: the mineral which underpins the operations of the primary employer, Columbus. It is therefore not clear to me that the strike at the plant of Manganese Metal could possibly have any direct or indirect effect on the business of Columbus. Even accepting that Manganese Metal is a wholly owned subsidiary of Samancor, it merely establishes a nexus between it and Columbus. But a mere nexus which does not have an effect on the primary employer’s business is insufficient to permit a secondary strike. As far as Samancor is concerned, it provides at least 80% of the chrome required by Columbus from its plant in Middelburg. At present Columbus is engaged in a shut-down or a maintenance programme which has the effect that its furnaces are inoperative or at least one of them is inoperative. It will recommence its operations on Monday, 26 July. It has four days’ worth of stock. Samancor is a partner in the joint venture with Columbus Stainless Steel. It has a stake in the outcome of the strike, for any further losses which Columbus may suffer will be debited, proportionally, to its account. It is therefore in a position to influence the business of Columbus 34 PILLAY AJ Billiton II.100 [78] If a relationship in which the secondary employer is able to influence market sentiment and ratings agencies that, in turn, impact adversely on the commodity price, the share price and the valuations of the primary employer, that would be another example. But the links must be proven. AMCU’s attempt at leading expert evidence in the Labour Court to prove this effect failed because the witness did not qualify as an expert. Furthermore, on the facts, the Labour Court found that so many variables influence share prices – ranging from “macro geopolitical issues to the appointment of a new chief executive” – that it could not pin down any fall off to the intended secondary strike alone.101 The evidence was lacking in that instance. Another consideration when assessing the impact of strikes is this: a drop in production, and concomitantly, labour costs induced by the “no work, no pay principle”, could increase demand and thereby spike the price of the commodity or goods – paradoxically, a most desirable consequence of a strike for the industry. if spurred on by a secondary strike at its plant and mines. The fact that it may have no say in the day–to-day running of Columbus is not especially pertinent. The strike is not about the day-to-day operations of Samancor. It is about the wages paid to its NUMSA employees. This is a matter of capital funding in which Samancor has a real and substantial interest and thus an incentive to use its influence on the collective bargaining process between Columbus and its NUMSA employees. Now that I have found that the secondary strike has a possible direct or indirect effect on the business of Columbus, I must consider whether the nature and extent of the strike at the applicants’ plants and mines was reasonable in relation to the effects of the strike at Columbus and on Columbus’s business. I do not think that it is necessary to weigh up the damage inflicted to the applicants against the effect of the strike on the business at Columbus but of course I do not ignore it. It seems to me that section 66(2)(c) requires me to concentrate on the nature and extent of the strike, that is the withholding of labour, its timing and other ramifications in relation to the effects which it may have on the business of Columbus. It seems to me that it would have been reasonable for the respondent to have targeted only the chrome ore mines and chrome or ferro alloy producing plants which belong to Samancor. Insofar as Manganese Metal is concerned, the secondary strike at its mines and plants will not have a possible effect on the business of Columbus. It is therefore clear to me that an interdict should be granted in regard to the operations of Manganese Metal.” 100 Billiton II above n 96 at para 18 states: “The conclusion I have reached above that this matter must be decided on the basis that the appellant does not own any shares in the primary employer - let alone being the majority shareholder - disposes of the respondent's main case. This is because the respondent’s real case was based on the assertion that the appellant is the majority shareholder in the primary employer and, because of that, will intervene in the primary dispute to protect its capital investment or capital funding.” 101 Labour Court judgment above n 2 at para 160. 35 PILLAY AJ [79] The complexity of relationships between entities in the private and public sectors militates against pinning down hard and fast “one size fits all” principles about what is proportional and reasonable. Furthermore, establishing cause and effect of secondary strikes would be challenging at the best of times.102 More so in motion proceedings, especially when they are brought urgently, as secondary strike interdicts usually are. This difficulty presented in Billiton II when the Labour Appeal Court had to have recourse to Plascon-Evans103 to assess the facts to which to apply section 66(2)(c).104 What the cases cited above show is that it is difficult, but not impossible, to prove the effect, if any, between a secondary strike and the business of the primary employer.105 In cases in which the courts found an effect, they refused to interdict the secondary strikes. The effect does not have to be equal “upon the businesses of the secondary and primary employers” as suggested in the second judgment106 but reasonable, as I elaborate below. [80] Fourth, the assessment of the effect is, unsurprisingly, directed at “the primary employer”. After all, a secondary strike is by definition “a strike, or conduct in 102 Id at paras 170-7. 103 Plascon-Evans Paints Ltd v van Riebeeck Paints (Pty) Ltd [1984] ZASCA 51; 1984 (3) SA 623 (A). 104 Billiton II above n 96 at para 24: “I am of the view that the version put up by the respondent is not one that is so untenable or that it so inherently improbable or far-fetched that it would be justified to simply reject it on the papers. For this reason, and applying the Plascon-Evans approach, it seems to me that the matter must be decided on the basis of the respondent’s version. That, therefore, means that it has to be accepted, for purposes of the determination of this appeal, that Samancor makes a product called ‘paste’ which it then sells to the appellant and the latter uses that product to make ‘electrodes’. The respondent's version is also that every week the ‘paste’ is loaded on to trucks ‘at Ferroveld’. However, the difficulty with this part of the respondent's case in this matter is that the information placed before the court is not sufficient to justify a conclusion that, based on the paste issue, the nature and extent of the proposed secondary strike may possibly be reasonable. No indication has been given as to the size or quantity of the paste that is transported from the primary employer to the appellant. No indication has been given as to how many workers employed by the appellant have something to do with the paste. The indication is that the paste is transported from the primary employer to the appellant weekly. On this information I am unable to conclude that the nature and extent of the secondary strike may possibly be reasonable. If I cannot reach such a conclusion, I cannot go on to find that the condition prescribed by section 66(2)(c) has been met. Since that condition has not been met, the proposed secondary strike would be unprotected.” 105 See for instance Samancor above n 88 and SALGA I above n 18. 106 See the second judgment at [155]. 36 PILLAY AJ contemplation or furtherance of a strike, that is in support of a strike by other employees against their employer”.107 [81] Fifth, the reference to the effect on the primary employer does not imply that the secondary employer is excluded altogether from the assessment. The impact on the secondary employer comes into the analysis in interpreting whether “the nature and extent of the secondary strike is reasonable in relation to the effect that the secondary strike may have”. [82] Adding the words “and the secondary employer” at the end of section 66(2)(c) would have fortified the interpretation that reasonableness also applies to the secondary employer. However, elevating the primary and secondary employers to the same level would have undermined the definition of secondary strikes which targets the primary employer. Importantly, it would have decisively and unambiguously shifted the balance favourably towards the secondary employer. This shift was unlikely to enchant the trade union parties negotiating the draft Bill. Omitting the words “and the secondary employer” may be explained as probable resistance from the trade union negotiators to their inclusion. [83] Assessing the reasonableness of the secondary strike on the secondary employer is not disloyalty to the text by subversively reading in the words “and the secondary employer” at the end of section 66(2)(c). Quite the contrary. The absence of these words informs the way in which reasonableness of the secondary strike on the secondary employer will be assessed. That is, not based on parity with the primary employer. This approach contextualises the interpretation of the LRA in its genesis as a negotiated draft Bill. [84] Irrespective of the reasons, omitting the words “and the secondary employer” at the end of section 66(2)(c) distinguishes the primary and secondary employers. In 107 Section 66(1) of the LRA. 37 PILLAY AJ relation to the primary employer, the secondary strike must have an effect. In relation to the secondary employer, the secondary strike must be proportional, even if it has an effect on the primary employer. Importantly, the omission imbues section 66(2)(c) with a degree of flexibility that invites judicial intervention to determine reasonableness case by case. [85] Reasonableness is not an abstraction. Instead, it harks back to its roots in the Constitution, to give effect to section 36 specifically. To be lawful, the secondary strike must be reasonably capable of having the possible effect. This begins, not with a value judgement, but with a sound enquiry into the facts. In the circumstances of Samancor, SALGA I and SALGA II, the courts decided, on the facts, that the secondary strike could influence the business of the primary employer.108 [86] Furthermore, “reasonable” also qualifies “the nature and extent” of the secondary strike. This does involve a value judgement. Conceptually, the phrase “the nature and extent” in section 66(2)(c) is cast widely so that it includes secondary strikes in all sectors and industries, private and public, and may take many forms. Factors to determine the nature and extent of the secondary strike would include the duration and form of the strike, that is, whether it is a go-slow, intermittent or stay-away strike for a day or longer; the number of employees involved; their membership of trade unions; their conduct, including whether it is peaceful or violent; and the sector(s) in which the primary and secondary strikes occur, that is, whether it is in the private or public sector, or the retail, agricultural, manufacturing, or mining sectors, and so on.109 [87] It follows that if a secondary strike is incapable of having any effect whatsoever on the business of a primary employer, that would not be reasonable. That would be the end of the enquiry. For, the most basic requirement to qualify for a lawful secondary strike will not have been met. Proving that the nature and extent of a secondary strike 108 SALGA II above n 11 at paras 10-6; SALGA I above n 18 at paras 21-2 and Samancor above n 88 at para 31. 109 Billiton II above n 96 at paras 20-4; SALGA I id at paras 17-9 and Samancor id at paras 29-31. 38 PILLAY AJ could possibly have an effect on the business of the primary employer requires evidence. Such evidence will be harder to establish if the link between the primary employer, and its workers, on the one hand, and the secondary employer, and its workers, on the other hand, is weak. After establishing that the secondary strike has an effect on the business of the primary employer, assessing the nature and effect of the secondary strike calls for a value judgement. [88] The phrase “in relation to” in section 66(2)(c) unambiguously anticipates a comparison. The comparators are firstly, the secondary strike and business of the primary employer, and secondly, the secondary strike and business of the secondary employer. The standard prescribed for comparison is reasonableness. Thus, the phrase “reasonable in relation to” imports proportionality in assessing reasonableness. Conceptually, proportionality and reasonableness often converge but not necessarily. In balancing the right to strike with the rights of primary and secondary employers, all factors must be considered to determine, not just proportionality, but ultimately, reasonableness. For reasonableness is the standard or attribute to which the LRA aspires for secondary strikes because the right to strike is constitutionally entrenched. Below, an example will illustrate an instance when proportionality and reasonableness can be in tension with each other and the right to strike. [89] Recognising proportionality and reasonableness of the secondary strike anticipates some safeguards for secondary employers. Otherwise, they would have no means of protecting their businesses. Their employees, who are non-unionised or who belong to trade unions other than the one calling for the secondary strike, may also want protection. Proportionality limits the right to participate in a secondary strike and introduces safeguards to rebalance the rights of secondary employers more favourably relative to the primary employer. [90] Assessing the impact of the secondary strike on the business of the primary employer is a question of fact determined case by case. Assuming, as the second judgment does, that secondary strikes “may have some impact but to a lesser degree” 39 PILLAY AJ than the primary strike,110 that will not be true in every instance. Much would depend, for instance, on the relationship between the primary and secondary employers, the number of employees involved in the strikes and whether the businesses of the primary and secondary employers are interrupted partially or totally. A partial shutdown of the business of a primary employer could be more greatly impacted by a total shutdown of the secondary employer who supplies components to the primary employer. On its own, this fact would discourage interference in the power-play by interdicting the secondary strike, unless after considering all the factors, some of which are mentioned above, the impact of the secondary strike on the secondary employer is not only disproportionate but also unreasonable. In making this assessment, the constitutionalising of the right to strike and to bargain collectively as the primary means of eliminating poverty and inequality must weigh in. [91] Proportionality and reasonableness enquiries invite judicial supervision of the interpretation and application of the law to the circumstances. Academic opinion favours “an approach that requires the harm caused to the secondary employer to be proportional to its likely impact on the business of the primary employer”.111 In the opinion of Seady and Thompson, the LRA leaves it “to the Labour Court to define the parameters of protected secondary strikes within the meaning of section 66(2)(c)”.112 Cooper concludes that in “the final analysis however, as in the foreign jurisdictions described above, the Labour Court will play a critical role in clearly defining the nature of this balance”.113 [92] The stance adopted in the second judgment of disavowing a proportionality assessment is excessively inclined towards protecting altruistic sympathy strikes. Trade unions and their members would have considerably more leverage. It would topple the equilibrium that section 66(2)(c) seeks to establish. 110 See the second judgment at [154]. 111 Seady and Thompson above n 60 at 328. 112 Id at 328. 113 Cooper above n 51 at 784. 40 PILLAY AJ [93] Proportionality assessments to determine reasonableness are necessary for reasons that relate to the parties engaged in secondary strikes, and beyond. Persons other than the employers and employees involved in the secondary strikes may be affected if their rights are violated. Jettisoning a proportionality assessment would exclude access to courts to invite judicial scrutiny when rights under the LRA, the Constitution and other laws could be implicated. [94] In the administrative law context, ouster clauses were notorious for concealing rights violations in our pre-democratic order.114 In the context of labour law, ousting the courts from assessing proportionality would distort the dispute system design of the LRA. The LRA anticipates and prescribes peaceful and appropriate forms of dispute resolution for almost every conceivable labour dispute. Allowing rights violations to burgeon without judicial scrutiny of secondary strikes would be out of kilter with that design and unwholesome for a constitutional democracy. And denying access to courts to have the proportionality of secondary strikes assessed could have dire consequences if those whose rights are violated have to resort to self-help to protect their interests. Effective management of conflict would be sacrificed. [95] For these reasons, I respectfully dissociate myself from the view espoused in the second judgment that section 66(2)(c) occludes a proportionality assessment. Interpretation applied [96] By recognising that section 66(2)(c) imports a proportionality test to the reasonableness standard within the provision, it follows that AMCU’s reliance on both Hextex and Billiton I is misplaced. Neither was followed in SALGA I.115 In fact, both 114 Hoexter “The Principle of Legality in South African Administrative Law” (2004) 4 Maquarie Law Journal 165 at 165-7 and Olayinka “Judicial Review of Ouster Clause Provisions in the 1999 Constitution: Lessons for Nigeria” (2018) 9 Nnamdi Azikiwe University Journal of International Law and Jurisprudence 138 at 151. 115 In SALGA I above n 18 at paras 14-5: “To the extent that recent decisions by this Court, for example, Billiton I and Hextex, respectively discount the effect of a secondary strike on the secondary employer as a relevant factor, and reject a proportionality requirement as fundamental to the application of 41 PILLAY AJ were effectively overruled by the Labour Appeal Court in Billiton II,116 SALGA II and now, in this case.117 In SALGA II, the application of a proportionality assessment was not debated because “quite properly, neither the appellant nor the union contest[ed] the proposition that section 66(2)(c) of the Act, imports a proportionality test”.118 The Labour Appeal Court went on to apply proportionality to the facts of that case. [97] Having established that section 66(2)(c) imports a proportionality analysis, I turn now to consider whether the secondary strikes in this matter were proportionate. As the matter is moot, I do so purely for the purposes of illustrating how the principles discussed above would apply to facts as in this case. [98] Economic loss is a predictable consequence of strikes generally. To give effect to a lower threshold for tolerance for loss, section 66(2)(c) imports proportionality and reasonableness. Neither standard limits primary strikes as defined under section 213 of the LRA. [99] The Labour Court found that the multi-employer secondary strikes would individually and collectively cause “irrecoverable” losses. Secondary strikes that wreak macro-economic havoc, bring industry to a standstill, jeopardise job security, cause commodity prices to collapse and investors to take flight, would render the secondary strikes unreasonable. Both the primary and the secondary employers would be affected. section 66(2), I respectfully disagree with that approach. An assessment of the nature and extent of the secondary strike clearly contemplates that its impact on the business of the secondary employer is a fundamental factor, and that an assessment of that impact is required. Further, the use of the words “reasonable in relation to” in section 66(2)(c), clearly import a proportionality assessment, not unlike the use of the term “reasonable” in section 36 of the Constitution in relation to the proportionality assessment of limitations on constitutional rights.” 116 Billiton II above n 96 at para 24. 117 Labour Court judgment above n 2 at paras 19-21. 118 SALGA II above n 11 at para 9: “In these proceedings, quite properly, neither the appellant nor the union contest the proposition that section 66(2)(c) of the Act, imports a proportionality test. What is required to be determined, as the court a quo correctly observed, is the reasonableness of the nature and extent of the secondary strike (which inevitably involves an enquiry into the effect of the strike on the secondary employer) in relation to the effect on the business of the primary employer (which inevitably involves an enquiry into the extent of the pressure placed on the primary employer).” 42 PILLAY AJ While the primary employer would have the option of resolving the bargaining dispute, the secondary employers may have no way of influencing that resolution. However, to succeed in securing an interdict, secondary employers must convince the courts that the secondary strikes are unreasonable. [100] In this instance, the secondary strikes were unreasonable because, firstly, they had no effect on the business of Sibanye. Secondly, the impact of the secondary strikes on the secondary employers, and beyond, was unreasonably destructive. AMCU’s plan to limit the secondary strikes to seven days did not mitigate the disproportionate impact of the secondary strikes. Neither did its suggestion that the secondary employers should have recourse to replacement labour. Aside from being anathema to trade unions at the best of times, suggesting replacement labour as a way of limiting the impact of the secondary strikes was more cynical than practical, considering that implementing it on seven days’ notice was not doable. [101] The findings that section 66(2)(c) encompasses proportionality as a standard to assess the reasonableness of secondary strikes, and that, on the facts, all secondary strikes in this instance had no effect on the business of the primary employer but would have had disproportionately devastating effects on the secondary employers, disposes of the primary issue for determination. [102] The first subsidiary question about whether the assessment of the effect on the business of the primary employer should be per individual secondary employer or collectively is answered with reference to the section. Section 66(2)(c) uses the singular “secondary strike”. As a general rule of legislative drafting, the singular includes the plural.119 Therefore, the binary posed in the subsidiary question is adequately covered in section 66(2)(c). In each case, the facts will indicate what effect, if any, secondary strikes, individually and collectively, have on the business of the primary employer. The answer lies in fact – that is, whether the secondary strike has an effect, rather than 119 Section 6(b) of the Interpretation Act 33 of 1957. See also Gcilitshana v General Accident Insurance Co SA Ltd 1985 (2) SA 367 (C). 43 PILLAY AJ in law. As stated above, every secondary strike must have an effect on the business of the primary employer, otherwise it would be automatically unlawful. On the facts, the Labour Court found that the secondary strikes, individually and collectively, had no effect on the business of the primary employer. [103] However, on principle the Labour Court interpreted a section 66(2)(c) enquiry into the reasonableness of the secondary strike to prohibit “the grouping together of a collection of secondary employers in a specific industry and assessing what the combined effect of a broader industry secondary strike would have on the primary employer”. In my view, nothing in section 66(2)(c) prohibits a “grouping together”, aggregation and multi-employer secondary strikes in principle. [104] If the concern about permitting the aggregation of secondary strikes is their scale and size, then prohibiting the grouping together is not a solution. For instance, a holding company with many subsidiaries may be the secondary employer in the same or multiple industries. Identifying and defining the secondary employer would be a source of unnecessary conflict when the real issue is the effect of the secondary strike on the business of the primary employer, irrespective of the identity and size of the employer. Similarly, a strike at a secondary employer that is a monopoly would also not address the issue of scale, size and impact on the secondary employer and the economy. [105] Section 66(2)(c) already offers two strong safeguards against any secondary strike – aggregated or individual. First, each secondary strike must have an effect on the business of the primary employer. Any secondary strike that does not have an effect would be unlawful. Second, secondary strikes individually and cumulatively must meet the standard of reasonableness to be lawful. [106] The Labour Appeal Court has used both controls – having an effect and reasonableness – discerningly. Samancor is the classic example of discerning the differences between secondary employers who do and those who do not have an effect. 44 PILLAY AJ Manganese Metals as a wholly owned subsidiary had a nexus with Samancor but because Manganese Metals provided no chrome, it had no effect on the business of the primary employer, Columbus. In contrast, Samancor had an effect because it was the supplier of 80% of the chrome required by Columbus. [107] These examples show that assessing effect and proportionality are intensely fact sensitive. What factors would determine whether there is an effect and whether it is proportional should be left to the courts to decide case by case. Because industries and services vary greatly, the factors to consider would also vary. I have mentioned some factors above. A proportionality assessment must also take account of whether the secondary strikes should be considered singly or cumulatively (grouped together). It is impossible in this judgment to forecast what effect secondary strikes will have in every instance to outlaw absolutely the grouping together of secondary strikes for the purposes of a proportionality assessment. However, it is possible, in a particular circumstance to limit the grouping together of secondary strikes. [108] The second subsidiary question, namely whether the prospect of violence is a consideration when assessing the reasonableness of secondary strikes, turns firstly on the definition of strikes. The “partial or complete concerted refusal to work, or the retardation or obstruction of work,” implies that, by definition, all strikes must be peaceful exercises of power to resolve disputes. This is reinforced by section 17 of the Constitution which guarantees that, “[e]veryone has the right, peacefully and unarmed, to assemble, to demonstrate, to picket and to present petitions”. Once strikes cease to be peaceful, they lose the protection of the law. Therefore, the prospect of violence during secondary strikes, properly diagnosed, would be a factor in assessing the reasonableness of a secondary strike. A history of violence accompanying strikes would found a proper diagnosis, as it did in this case. However, if an interdict to stop the violence is possible without interfering with the secondary strike, then that should be the preferred route. 45 PILLAY AJ Conclusion [109] This matter raises a discrete question of law and two subsidiary issues. Primarily, does section 66(2)(c) import the principle of proportionality in assessing the reasonableness and the substantive lawfulness of secondary strikes? Our response is “yes”. Section 66(2)(c) seeks to balance the impact of secondary strikes on secondary employers, on the one hand, with their effect on the business of the primary employer on the other hand. Thus, on the one extreme, if secondary strikes have no effect on the primary employer, or, on the other extreme, if the effect is disproportionately harsh on secondary employers, they would be entitled to interdict the strike under section 66(3). This principle of proportionality derives not only from the reasonableness requirement to be found in section 66(2)(c), but also from the Constitution and international law. [110] The general rule that costs in labour matters do not follow the result applies.120 The central issue is a matter of mutual interest to both parties and beyond. Each party should pay its own costs. The rule also applies to reverse the cost orders in the Labour Court (in refusing leave to appeal), and the Labour Appeal Court. Order [111] The following order is made: (a) (b) Leave to appeal is granted. Save as set out below, the appeal against the judgments of the Labour Court and the Labour Appeal Court is dismissed. (c) The costs orders in the Labour Court and the Labour Appeal Court are set aside. (d) Each party shall pay its own costs in the Labour Court, the Labour Appeal Court and this Court. 120 Union for Police, Security and Corrections Organisation v South African Custodial Management (Pty) Ltd [2021] ZACC 26; 2021 JDR 2220 (CC). 46 JAFTA J JAFTA J: [112] I have had the pleasure of reading the judgment by my colleague Pillay AJ (first judgment). While I agree that leave should be granted and that the costs order issued by the court below ought to be set aside, I hold that the appeal on the interpretation of section 66(2)(c) should have succeeded. I also agree that leave should be granted despite the fact that the matter has become moot. [113] AMCU have sought leave to appeal against the decision of the Labour Appeal Court. The Labour Appeal Court had dismissed AMCU’s appeal against the decision of the Labour Court in terms of which the secondary strike in which AMCU was to participate was declared to be unprotected. This declaration was based on the interpretation of section 66(2)(c) by the Labour Court. [114] By the time the appeal was heard by the Labour Appeal Court, the matter had become moot in that AMCU was no longer pursuing secondary strikes. The primary strike had already been resolved and the need for secondary strikes had fallen away. Based on this reason alone, the Labour Appeal Court decided to dismiss the appeal. Background [115] AMCU commenced a primary strike at various mines owned by Sibanye in November 2018. That strike continued until April 2019 when it was resolved. Meanwhile in February 2019 AMCU issued notices for secondary strikes on various gold mining companies. These were intended to commence on 28 February 2019. Having received notices, the affected companies instituted, as a matter of urgency, applications against AMCU in the Labour Court. They all sought an interdict restraining AMCU from going ahead with its secondary strikes. At the heart of the 47 JAFTA J dispute was whether the requirements of a secondary strike set out in section 66(2) of the LRA were satisfied. [116] This arose because the mining companies contended that the secondary strikes would have no impact on the businesses of Sibanye, the primary employer. They argued that the secondary strikes would bring their businesses to a complete halt and, as a result, the strikes would have a devastating effect on those businesses. They also pointed out that the primary strike at Sibanye was accompanied by unlawful acts of violence and that they sought to prevent violent damage to property and loss of life and limb. [117] The applications by the mining companies were opposed by AMCU which argued that all requirements of section 66(2) were met. But the Labour Court held that the reasonableness requirement in section 66(2)(c) was not satisfied because the secondary strikes would cause more harm to the businesses of the secondary employers than to Sibanye’s business. Accordingly, the Labour Court declared that each secondary strike was unprotected and made no order as to costs. [118] Since the primary strike was resolved, the matter has become moot. The question that arises is whether that mootness is a bar to the entertainment of this appeal. Our jurisprudence clearly shows that mootness is not an absolute bar to deciding an appeal. Our courts have a discretion to exercise judicially and entertain an appeal that is moot under certain circumstances.121 A court may refuse to hear a moot appeal if its decision would be academic and not serve any other purpose. But if the decision of the appeal court would be useful to the parties in the future or even to other parties, the court may decide to entertain the appeal. This may be the case where a matter involves the interpretation of a statute that continues to operate. Here, the interpretation of section 66(2) of the LRA would provide useful guidance in the future if unions wish to embark on secondary strikes. Employers and unions would know the exact requirements of the provision. 121 Pillay above n 34 and Independent Electoral Commission above n 34. 48 JAFTA J [119] This Court in the past has entertained a moot appeal involving the construction of the provisions of the LRA for similar reasons. In POPCRU this Court said: “The meaning given to this section will have a practical effect on all future disputes involving agreements that declare thresholds of representativeness. This illustrates that a decision of this Court on the merits will be of great benefit to workers, trade unions and employers in the future. This is because section 18 of the LRA continues to apply to their relationships and the Act was enacted to give effect to rights in section 23 of the Constitution.”122 [120] For these reasons I agree that the standard for deciding a moot appeal is satisfied here. The relevant provision continues to be in force and there is a likelihood of a trade union embarking on a secondary strike in the future, in order to maximise pressure on the primary employer to return to the bargaining table and break the deadlock that led to the primary strike in the first place. This paves the way to construing section 66(2). Proper approach to interpreting the LRA [121] The starting point must be the LRA itself because it tells us how it should be interpreted. Section 3 of the LRA obliges any person who applies the Act, including a court, to interpret it in compliance with the Constitution and public international law obligations of the Republic and to give effect to the LRA’s primary objects.123 In simple terms this injunction notably means that whenever the LRA is interpreted, it must be given a purposive meaning which complies with both the Constitution and international law. For example, where the provision under construction is reasonably capable of two meanings, one of which is compliant with the Constitution and international law 122 POPCRU above n 35 at para 72. 123 Section 3 of the LRA provides: “Any person applying this Act must interpret its provisions— (a) (b) (c) to give effect to its primary objects; in compliance with the Constitution; and in compliance with the public international law obligations of the Republic.” 49 JAFTA J obligations of the Republic, the interpreting court must prefer the meaning that is consistent with the Constitution and international law over the one that is not. [122] Moreover, section 1 of the LRA sets out its primary objects to which effect must be given through interpretation.124 Of importance for present purposes among those primary objects is giving effect to the labour rights entrenched in section 23 of the Constitution which is a successor of section 27 of the interim Constitution mentioned in section 1. The other primary object is ensuring compliance with the Republic’s obligations as a member of the ILO. Here the purposive meaning given in section 66(2) must advance these two purposes. I will return to this point in detail later. [123] To a degree, the injunction in section 3 of the LRA echoes the obligation imposed by section 39(2) of the Constitution.125 This provision obliges every court when interpreting legislation to “promote the spirit, purport and the objects of the Bill of Rights”. Evidently here, the right which must be advanced is the one regulated by section 66(2) and that is the right to strike. At an interpretation level this means that if 124 Section 1 of the LRA provides: “The purpose of this Act is to advance economic development, social justice, labour peace and the democratisation of the workplace by fulfilling the primary objects of this Act, which are— (a) (b) (c) to give effect to and regulate the fundamental rights conferred by section 23 of the Constitution of the Republic of South Africa, 1996; to give effect to obligations incurred by the Republic as a member state of the International Labour Organisation; to provide a framework within which employees and their trade unions, employers and employers’ organisations can— (i) collectively bargain to determine wages, terms and conditions of employment and other matters of mutual interest; and (ii) formulate industrial policy; and (d) to promote— (i) (ii) (iii) (iv) orderly collective bargaining; collective bargaining at sectoral level; employee participation in decision-making in the workplace; and the effective resolution of labour disputes.” 125 Section 39(2) of the Constitution provides: “When interpreting any legislation, and when developing the common law or customary law, every court, tribunal or forum must promote the spirit, purport and objects of the Bill of Rights.” 50 section 66(2) is reasonably capable of a construction that promotes the right to strike, that interpretation must be preferred over the other meanings which do not. In Makate this Court pronounced: JAFTA J “The objects of the Bill of Rights are promoted by, where the provision is capable of more than one meaning, adopting a meaning that does not limit a right in the Bill of Rights. If the provision is not only capable of a construction that avoids limiting rights in the Bill of Rights but also bears a meaning that promotes those rights, the court is obliged to prefer the latter meaning.”126 [124] In addition, a reminder on what the process of interpretation entails is necessary before we undertake that task. Interpretation of a statute constitutes a process of determining what Parliament meant in a particular provision by attaching a meaning to the words used in the provision in question. Unless defined, those words must bear their ordinary meaning and the process relates to assigning meaning to each and every word that appears in the provision. This means that words which do not appear in the provision cannot be added to the provision under the guise of interpretation. For the importation of words into a statutory provision falls outside the scope of interpretation. In other words, reading into a statute words which were not included by the lawmaker is not part of the interpretation process. In light of what occurred here, it bears emphasis that interpretation is limited to attributing meaning to words used in a provision, and that does not incorporate the importation of words which we think should have been included but were omitted. [125] This is because interpretation is not part of law making. It does not entail adding words to a provision or subtracting them. It is not the task of an interpreter to alter a provision enacted by Parliament under the rubric of interpretation. The fact that it is competent for a court under certain circumstances to read words into a statute to cure a constitutional defect does not mean that the court may do so while interpreting the provision. Therefore, as a matter of interpretation, words which do not appear in 126 Makate v Vodacom Ltd [2016] ZACC 13; 2016 (4) SA 121 (CC); 2016 (6) BCLR 709 (CC) at para 89. 51 JAFTA J section 66(2)(c) cannot be incorporated into that provision so as to give it a meaning which we think it ought to carry. [126] Fidelity to the LRA demands that we should interpret section 66(2) in the exact form in which it was enacted, without any changes. The principle of respecting the language chosen by the lawmaker is entrenched in the jurisprudence of this Court. First, the injunction in section 39(2) may be discharged only if the language used in the legislation is reasonably capable of a meaning that promotes the objects of the Bill of Rights. This is consistent with the principle that if “the language used by the law giver is ignored in favour of a general resort to ‘values’ the result is not interpretation but divination”.127 [127] In Bertie Van Zyl this Court affirmed the importance of maintaining fidelity to the language of the lawmaker. There the Court declared: “A contextual or purposive reading of a statute must of course remain faithful to the actual wording of the statute. When confronted with legislation which includes wording not capable of sustaining an interpretation that would render it constitutionally compliant, courts are required, as discussed above, to declare the legislation unconstitutional and invalid.”128 [128] Having outlined in detail the backdrop against which section 66(2) must be construed, it is now convenient to interpret the provision. But before doing so, I must point out that the Labour Court here did not pay attention to the actual text of the provision. After a comprehensive analysis of the facts, that Court proceeded straight to a conclusion it reached on its understanding of the provision. The Court held: “The reasonableness requirement involves a proportionality assessment: the harm caused by the secondary strike to the secondary employer may be in proportion to the harm potentially caused to the primary employer as a consequence of the secondary 127 S v Zuma [1995] ZACC 1; 1995 (2) SA 642 (CC); 1995 (4) BCLR 401 (CC) (Zuma) at para 18. 128 Bertie Van Zyl above n 81 at para 22. 52 JAFTA J strike. In my view, this calls for an assessment of three key issues: the effect of the secondary strike on the secondary employer, the possible direct or indirect effect that the secondary strike may have on the business of the primary employer and the proportionality of the harm caused to the primary and secondary employers respectively.”129 [129] Proceeding from this premise, the Labour Court considered seriatim (separately) what it identified as the key issues. These were the effect of the secondary strike on the secondary employer, the proportionality of harm caused and the effect of the secondary strike on the business of the primary employer. Without stating the source of the first two issues, the Labour Court continued to consider the impact of each on the businesses of the secondary employer. That Court also considered the impact the secondary strikes had on the business of Sibanye, the primary employer. Notably, that Court focused on whether there was a link between the businesses of the secondary employers and that of the primary employer. On the facts the Labour Court concluded that the secondary strikes would have had no impact on the business of the primary employer. This was the ratio for the order issued. Interpretation of section 66 [130] Section 66(2) must be read and understood in the context of the whole section and as a result it is necessary to quote the entire provision. It provides: “(1) In this section ‘secondary strike’ means a strike, or conduct in contemplation or furtherance of a strike, that is in support of a strike by other employees against their employer but does not include a strike in pursuit of a demand that has been referred to a council if the striking employees, employed within the registered scope of that council, have a material interest in that demand. (2) No person may take part in a secondary strike unless— (a) the strike that is to be supported complies with the provisions of sections 64 and 65; 129 Labour Court judgment above n 2 at para 178. 53 JAFTA J (b) the employer of the employees taking part in the secondary strike or, where appropriate, the employers’ organisation of which that employer is a member, has received written notice of the proposed secondary strike at least seven days prior to its commencement; and (c) the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer. (3) Subject to section 68(2) and (3), a secondary employer may apply to the Labour Court for an interdict to prohibit or limit a secondary strike that contravenes subsection (2). (4) Any person who is a party to proceedings in terms of subsection (3), or the Labour Court, may request the Commission to conduct an urgent investigation to assist the Court to determine whether the requirements of subsection (2)(c) have been met. (5) On receipt of a request made in terms of subsection (4), the Commission must appoint a suitably qualified person to conduct the investigation, and then submit, as soon as possible, a report to the Labour Court. (6) The Labour Court must take account of the Commission’s report in terms of subsection (5) before making an order.” [131] In view of the interpretation given to section 66(2)(c) by the courts below and the first judgment, it is necessary to carefully read the entire section to determine whether that interpretation is correct and whether it accords with the injunctions in section 39(2) of the Constitution and section 3 of the LRA. Subsection (1) tells us what a secondary strike is. It says a secondary strike is a strike in support of a strike by other employees against their employer. Before workers can embark on a secondary strike against their own employer, there must be a strike by other employees against those employees’ employer. That is, their strike must be secondary to the first strike which is defined as the primary strike. However, the secondary strike must not only sequentially follow the primary strike, it must also support it. But importantly subsection (1) does not define the level of support. Therefore for as long as there is 54 JAFTA J support of the primary strike in the secondary strike, the definitional elements of a secondary strike are satisfied. [132] Subsection (2) prescribes the requirements which must be met by workers before they can embark on a secondary strike. But I propose to consider briefly the other subsections before examining in detail the text of subsection (2), because it must be read in the context of the whole section. Subsection (3) confers upon the employer of workers who contemplate a secondary strike, the right to apply for an interdict restraining the contemplated strike. This right is limited to and must be based on a contravention of subsection (2). And procedurally the exercise of this right is subject to section 68(2) and (3). Section 68(2) empowers the Labour Court to refuse the interdict if 48 hours’ notice of the application for the interdict is not given to the workers and other parties cited as the respondents. However the Court may permit a shorter notice period if good cause is shown and the respondents have been given written notice that affords them a reasonable opportunity to be heard. This means that an application for an interdict cannot be made ex parte. [133] For its part section 68(3) fixes a longer notice period where the notice for the impending secondary strike was given in writing at least 10 days before the commencement of the proposed strike. In that event, the secondary employer must give notice of at least five days to the respondents. This mandatory requirement of a notice of no less than five days overrides the discretion conferred on the Labour Court by section 68(2) and the shorter notice periods in that subsection. This means that where written notice of at least 10 days is given before the commencement of the secondary strike, section 68(2) is displaced by section 68(3), and the latter subsection applies exclusively. [134] Section 66(4), (5) and (6) regulate an investigation into whether the requirements of subsection (2)(c) have been met before the secondary strike begins. Subsection (4) authorises the parties to the interdict proceedings to request the CCMA to conduct an investigation into compliance with subsection (2)(c). Subsection (5) 55 JAFTA J obliges the Court, upon receipt of such request, to appoint a commissioner to conduct an investigation as a matter of urgency and submit a report to the Labour Court. Lastly, subsection (6) obliges the Labour Court to have regard to that report before granting the interdict. This is the scheme within which section 66(2)(c) must be read. Meaning of section 66(2) [135] Returning to the text of section 66(2), this provision is framed in the form of a prohibition. Its opening words are “no person may take part in a secondary strike unless . . .”. Then the provision lists three requirements which must be satisfied before the secondary strike begins. The first requirement is that the primary strike to be supported by the secondary strike must have complied with sections 64 and 65 of the LRA. In general terms this means that the primary strike itself must be protected as it complied with all necessary requirements prescribed by the LRA. [136] The second requirement relates to notice. Section 66(2)(b) proclaims that the workers contemplating to participate in a secondary strike must give the secondary employer written notice of at least seven days before the strike in question starts. This notice need not specify which workers will participate in the secondary strike because section 66(2)(b) does not require that specificity. Instead, it merely requires that the secondary employer receive a written notice of the proposed strike, of at least seven days before it begins. [137] With regard to what a notice of a strike must specify as envisaged in section 64(1)(b), the Supreme Court of Appeal in Moloto (SCA)130 had preferred a construction that required the notice to specify the number of workers who intended to participate in the strike so that the employer could make necessary arrangements in its business. The Court held the view that the purpose of giving notice was to enable the employer to make an informed decision on whether to yield to the workers’ demands 130 Equity Aviation Services (Pty) Ltd v SATAWU [2011] ZASCA 232; 2012 (2) SA 177 (SCA) (Moloto (SCA)). 56 JAFTA J or resist the strike by making necessary arrangements to protect the operation of its business against the strike. [138] A majority in this Court rejected that interpretation on the basis that section 64(1)(b) of the LRA does not expressly say notice must be given by every worker or their unions. On that construction, if one union has given notice of a strike, the other unions and workers may simply join the strike without notifying the employer. The majority reasoned: “The right to strike is protected as a fundamental right in the Constitution without any express limitation. Constitutional rights conferred without express limitation should not be cut down by reading implicit limitations into them, and when legislative provisions limit or intrude upon those rights they should be interpreted in a manner least restrictive of the right if the text is reasonably capable of bearing that meaning. The procedural preconditions and substantive limitations of the right to strike in the Act contain no express requirement that every employee who intends to participate in a protected strike must personally or through a representative give notice of the commencement of the intended strike, nor that the notice must indicate who will take part in the strike.”131 [139] Since this is a decision of this Court, we are obliged to follow its approach in interpreting provisions that limit the exercise of the right to strike by prescribing requirements. Section 64, just like section 66, prescribes requirements for participating in a protected strike. In view of the injunction in section 39(2) of the Constitution, we are bound to adhere to the approach laid down by the majority in Moloto. This is because if a provision like section 66(2) of the LRA trenches upon the right to strike, it must be given a meaning that imposes the least limitation of that right. [140] The third requirement of section 66(2) is to be found in (c) which reads: 131 SATAWU v Moloto and Another N.N.O. [2012] ZACC 19; 2012 (6) SA 249 (CC); 2012 (11) BCLR 1177 (CC) (Moloto) at para 44. 57 JAFTA J “No person may take part in a secondary strike unless— . . . (c) the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer.” [141] Textually, the provision refers to the secondary strike and its relation to the business of the primary employer. This ties up with the purpose of such strike which is support to workers who are involved in the primary strike. Section 66(2)(c) does not expressly or impliedly refer to the business of the secondary employer, let alone the relationship between that strike and the business of the secondary employer. It is completely silent on the impact the secondary strike may have on the business of the secondary employer. It bears emphasis that not a single word in section 66(2)(c) may be construed as meaning the business of the secondary employer. Nor is there a word in it that reasonably means that the secondary strike should have an impact on the business of the secondary employer. For this meaning to arise one has to read into the provision words like “the secondary strike must have an impact on the business of the secondary employer”. This cannot be done on the pretext of interpreting the provision. It is impermissible and at odds with the jurisprudence of this Court in decisions like Zuma and Bertie Van Zyl which laid down the principle that the language chosen by the lawmaker must be respected and that the interpretation must remain faithful to that language. [142] It is true that considerations of fairness and equity may forcefully pull one in the direction of seeking to protect the innocent secondary employer which has nothing to do with the dispute between the primary employer and its workers who are on strike. But for a number of reasons a court should not surrender to this temptation. The language of section 66(2)(c) is clear and unambiguous on what it requires. It requires the nature and scope of the secondary strike to be reasonable. It expressly says so. And it proceeds to inform us that the reasonableness of the secondary strike must be “in 58 JAFTA J relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer”. [143] The words “the secondary strike is reasonable in relation to the possible direct or indirect effect” plainly link the reasonableness of the secondary strike to the impact that strike may have on the business of the primary employer. There is no room for reading those words as saying that the secondary strike must have an effect on the business of the secondary employer, as the Labour Court has found here.132 And that Court failed to point out any words in the provision which carry that meaning. There is simply none. [144] This Court has, in accordance with the principles that the lawmaker’s language must be respected and fidelity to it is to be maintained, rejected the resort to the values of fairness and equity when interpreting a statute. In Cool Ideas the majority stated: “Cool Ideas contended that it would be inequitable for Ms Hubbard to be absolved from complying with the arbitrator’s award and from paying the outstanding approximately R550 000 due to Cool Ideas. I am of the view that equity considerations do not apply. But even if they do, as my colleague Froneman J suggests, the law cannot countenance a situation where, on a case-by-case basis, equity and fairness considerations are invoked to circumvent and subvert the plain meaning of a statutory provision which is rationally connected to the legitimate purpose it seeks to achieve, as is the case here. To do so would be to undermine one of the essential fundamentals of the rule of law, namely the principle of legality.”133 [145] Equally there are no words that bear “the proportionality of the harm caused to the primary and secondary employers, respectively”,134 as held by the Labour Court. No words at all, and the Labour Court also referred to none. This is plainly an importation of a requirement not prescribed by the lawmaker. It is contrary to the 132 Labour Court judgment above n 2 at paras 178 and 180. 133 Cool Ideas 1186 CC v Hubbard [2014] ZACC 16; 2014 (4) SA 474 (CC); 2014 (8) BCLR 869 (CC) (Cool Ideas) at para 52. 134 Labour Court judgment above n 2 at para 178. 59 JAFTA J principle that constitutional rights conferred without express limitation should not be cut down by reading implicit limitations into them. Here section 66(2)(c) does not expressly limit the right to strike by imposing a proportionality standard for determining the harm caused by the secondary strike upon the businesses of the primary and secondary employers. As Moloto informs us, the right to strike cannot “be cut down by reading implicit limitations” into it. [146] On the contrary, the express limitations in section 66(2)(a), (b) and (c) must be interpreted restrictively so as to have the least intrusion into the right to strike. This principle was also affirmed in SAPS where this Court held: “In order to ascertain the meaning of essential service, regard must be had to the purpose of the legislation and the context in which the phrase appears. An important purpose of the LRA is to give effect to the right to strike entrenched in s 23(2)(c) of the Constitution. The interpretive process must give effect to this purpose within the other purposes of the LRA as set out in s 1(a). The provisions in question must thus not be construed in isolation, but in the context of the other provisions in the LRA and the SAPS Act. For this reason, a restrictive interpretation of essential service must, if possible, be adopted so as to avoid impermissibly limiting the right to strike. Were legislation to define essential service too broadly, this would impermissibly limit the right to strike.”135 [147] Our jurisprudence underpinned by the various principles outlined above, recognises the centrality of the right to strike to the process of collective bargaining,136 hence the need for the link between the secondary strike and the business of the primary employer. In Bader Bop this Court observed: “This case concerns the right to strike. That right is of both historical and contemporaneous significance. In the first place, it is of importance for the dignity of workers who in our constitutional order may not be treated as coerced employees. 135 South African Police Service v Police and Prisons Civil Rights Union [2011] ZACC 21; 2011 (6) SA 1 (CC); 2011 (9) BCLR 992 (CC) (SAPS) at para 30. 136 Bader Bop above n 1 at para 13. 60 Secondly, it is through industrial action that workers are able to assert bargaining power in industrial relations. The right to strike is an important component of a successful collective bargaining system.”137 JAFTA J [148] It is evident from this jurisprudence that in a number of decisions this Court has recognised that the purpose of the right to strike is to promote collective bargaining, which in turn is one of the primary objects of the LRA. Therefore an interpretation that favours the right to strike gives effect to this primary object of the LRA. It also gives effect to compliance with the constitutional injunction of promoting the objects of the Bill of Rights. [149] Properly construed, the purpose of section 66(2) is to regulate the exercise of the right to participate in a secondary strike. In explicit terms the provision tells us that workers may participate in a secondary strike if three express requirements are met. Section 66(2)(c) makes it known that the secondary strike must be related to the business of the primary employer. The reason for this is simply that the secondary strike itself must have the potential of driving the primary employer back to the bargaining table. The impact the strike might have on the business of the secondary employer is of no relevance to the bargaining process between the primary employer and its workers. Therefore, this impact is not linked to the purpose of the secondary strike and that is why section 66(2)(c) does not refer at all to the business of the secondary employer and any impact the secondary strike might have on it. [150] Accordingly section 66(2)(c) serves one purpose only. This is the pressure the secondary strike must bring to bear on the business of the primary employer, plain and simple. The provision does not even remotely seek to protect the interests of the secondary employer. That is not its purpose and consequently it cannot be given a meaning that protects the business of the secondary employer. Moreover, it bears repetition that such an interpretation is not borne out by the language used in the text and is inconsistent with a number of principles relevant to the interpretation of 137 Id. 61 JAFTA J section 66(2) which imposes limitations on the right to strike. As observed in Moloto, the intrusion by section 66(2)(c) into the right to strike must only be as much as is necessary to achieve its purpose of establishing the link between the secondary strike and the business of the primary employer.138 [151] The required effect or impact need not be direct. An indirect effect suffices. Nor is it required to be actual. The possibility, and not probability, of such effect is good enough for purposes of meeting the requirements of section 66(2)(c). This means that for the workers of the secondary employer to participate in a secondary strike, they need to show that the strike would possibly have an impact on the business of the primary employer. A mere possibility does not guarantee that the effect would materialise. It may occur or it may not, and yet Parliament considered that to be sufficient to justify a secondary strike, provided that the other requirements in (a) and (b) are also met. [152] The sufficiency of the mere possibility of an effect cannot be reconciled with the proportionality requirement added by the Labour Court. This is because this standard requires that the harm caused to the business of the secondary employer must be proportionate to the harm the same strike would have on the business of the primary employer. It is inconceivable that the secondary strike may give rise to mere possible harm on the business of the secondary employer. If the employees of the secondary employer are on strike, it would mean that their work is not performed and that would be the actual harm caused to the secondary employer. If the entire workforce of the secondary employer embarks on a strike, its business would be brought to a complete halt unless temporary workers are engaged to replace the striking workers. And yet section 66(2)(c) requires proof of possible harm on the business of the primary employer which harm may be direct or indirect. 138 Moloto above n 131 at para 55. 62 JAFTA J [153] But more importantly the language used in the provision does not require that the harm on the business of the primary employer must be proportionate to the harm suffered by the business of the secondary employer. If Parliament wanted to include proportionality as one of the provision’s requirements, it could easily have done so. It did not. In addition, Parliament could not have used in the provision the type of language that is irreconcilable with the proportionality analysis if it had contemplated that proportionality should apply. [154] The proportionality standard is not only unsupported by the language of section 66(2)(c) but is also based on a mistaken assumption that a secondary strike may have an effect that is proportionate to both the secondary and primary businesses. This is not likely. The withdrawal of labour by employees of the secondary employer can hardly have a corresponding effect in size and impact on the business of the primary employer, because the workers who are involved in a secondary strike do not work for the primary employer. The greater impact upon the business of the primary employer is caused by the primary strike. This does not mean that the secondary strike can have no effect on the business of the primary employer. It may have some impact but to a lesser degree. [155] To hold otherwise would mean that secondary strikes would be permitted only on the rare occasions, if ever there were such occasions, where the secondary strike is shown to have an equal effect upon the businesses of the secondary and primary employers. That is why to date there is no single case where the standard was applied. The facts of many cases decided by the Labour Courts have not met this almost impossible test. [156] It is self-evident from the text of section 66(2)(c) that the effect the secondary strike may have on both businesses would most of the time differ. Take for example, the fact that the effect of the secondary strike on the business of the secondary employer would always be direct, whereas the provision permits an indirect effect on the business 63 JAFTA J of the primary employer. There can be no proportionate effect in that case, yet the requirements of section 66(2)(c) would be met. [157] I conclude that section 66(2)(c) requires the secondary strike to be reasonable only in relation to the business of the primary employer. It would be reasonable if it has a possible direct or indirect effect on the business of the primary employer. There is no need for such effect to compare in size with the effect that the strike has on the business of the secondary employer. The purpose of the section is not to protect the secondary employer but to have pressure brought to bear on the primary employer to either surrender to its employees’ demands or return to the bargaining table. The secondary employer may not surrender to those demands or bargain on them. [158] This interpretation of section 66(2)(c) is consonant with section 39(2) of the Constitution and the demands of section 3 of the LRA. It is also in line with the principles set out in Moloto where a similar provision was interpreted by this Court. In addition, the interpretation accords with the ILO’s position to the effect that secondary strikes should be subject only to the requirement that the primary strike be lawful.139 [159] Here the Labour Court and the decisions on which it relied failed to follow decisions of this Court in cases like Moloto which interpreted provisions of the LRA that imposed limitations on the right to strike. That Court also departed from its own previous decisions in cases like Billiton Aluminium.140 Had the Labour Court referred to relevant decisions of this Court, it could have arrived at a different conclusion. [160] However, since the merits have become moot I agree that the appeal may succeed only to the extent of the costs order made by the Labour Appeal Court. 139 First judgment at [44]. 140 Billiton I above n 29. 64 THERON J THERON J: [161] I have had the benefit of reading the first and second judgments. [162] My Sister Pillay AJ interprets the Labour Court to have held that section 66(2)(c) prohibits multiple secondary strikes in support of the same primary strike from simultaneously taking place at distinct places of employment (so-called multi-employer secondary strikes). [163] With respect, this misconstrues the findings of the Labour Court. Neither that Court, the Labour Appeal Court, nor any of the parties, have suggested that section 66(2)(c) means that multi-employer secondary strikes are prohibited. Quite simply, this was never in issue in this matter. Instead, the relevant question is: when a court assesses whether a secondary strike is reasonable in relation to its effects on the business of the primary employer, can it consider only one secondary strike, or many? In other words, can a court aggregate secondary employers for purposes of the section 66(2)(c) inquiry? The Labour Court answered this question in the following term “In my view, this factual inquiry [mandated by section 66(2)(c)] does not permit the grouping together of a collection of secondary employers in a specific industry and assessing what the combined effect . . . would have on the primary employer. Such an approach will ignore the critical question namely: the effect the secondary strike may have on the business of the primary employer in relation to the secondary employer, which calls for a consideration of the facts and an assessment of factors relevant to each secondary employer. To do differently would deprive each single secondary employer of the protection afforded to them by section 66(2)(c) of the LRA.” (Emphasis added.)141 141 Labour Court judgment above n 2 at para 22. 65 THERON J [164] The Labour Court held further that: “It appears that AMCU’s position is that the proportionality analysis is to be undertaken collectively vis-à-vis all the employers who are subject to a secondary strike at once, as opposed to an individual analysis to be undertaken vis-à-vis each employer which is subject to the secondary strike. I have already held that the enquiry into the reasonableness of the secondary strike does not permit the grouping together of a collection of secondary employers in a specific industry and assessing what the combined effect of a broader industry secondary strike would have on the primary employer. The aforesaid assessment is to be undertaken in respect of the proportional harm done to each of the applicants and must be assessed individually.”142 (Emphasis added.) [165] And, on this score, the Labour Appeal Court held: “Whilst the facts of the SALGA case were different to that which applied in the present dispute, this Court, in confirming the judgment of Van Niekerk J, clearly established a test that the mandated enquiry has to focus on the individual secondary employer. There is, therefore, no reason to entertain an appeal with regard to a point of law which has already been settled by this Court.”143 (Emphasis added.) [166] Both the Labour Court and Labour Appeal Court were clear: for purposes of the section 66(2)(c) inquiry, a court is not permitted to aggregate secondary strikes.144 Neither court held that multi-employer secondary strikes are prohibited by section 66(2)(c). [167] To the extent that my Sister Pillay AJ diverges from these findings of the Labour Court and Labour Appeal Court, I disagree. First, my Sister’s finding flies in the face of Labour Appeal Court authority. In SALGA, the Labour Appeal Court held 142 Id at paras 147-8. 143 Labour Appeal Court judgment above n 2 at para 21. 144 Id and Labour Court judgment above n 2 at para 22. 66 THERON J that section 66(2)(c) requires that a court “weigh the effect of the secondary strike on the secondary employer and the effect of the nature and extent of the secondary strike on the business of the primary employer.”145 Respectfully, the first judgment’s departure from this authority is premised on a misunderstanding of what the Labour Court held in respect of aggregation. The Labour Court did not hold that multi- employer secondary strikes are prohibited. It held that, for purposes of the section 66(2)(c) inquiry, a court cannot aggregate secondary strikes. [168] Second, this conclusion is fortified by the text, context and purpose of section 66(2)(c). That section provides that “[n]o person may take part in a secondary strike unless . . . the nature and extent of the secondary strike is reasonable in relation to the possible direct or indirect effect that the secondary strike may have on the business of the primary employer”. (Emphasis added.) While it is true that the general rule is that the singular includes the plural, this rule is subject to the caveat that it only applies unless the context indicates otherwise.146 And here, the context makes clear that the use of the singular “secondary strike” was deliberate and precludes an aggregation of secondary strikes for purposes of the section 66(2)(c) inquiry. This is because section 66(3) provides that “a secondary employer may apply to the Labour Court for an interdict to prohibit or limit a secondary strike that contravenes subsection (2)”. Plainly, such an employer has no standing to interdict a secondary strike at a workplace to which that employer has no connection. Thus, for purposes of section 66(3), “secondary strike” must refer to an individual secondary strike, and not to an aggregation of secondary strikes. As a result, to hold that “secondary strike” as it appears in section 66(2)(c) can refer to an aggregation of secondary strikes requires that a different meaning is ascribed to “secondary strike” in sections 66(2)(c) and section 66(3). Trite principles of statutory interpretation tell us that such an interpretation is to be avoided.147 145 SALGA II above n 11 at para 10. (Emphasis added.) 146 Section 6(b) of the Interpretation Act. 147 Head of Department, Mpumalanga Department of Education v Hoërskool Ermelo [2009] ZACC 32; 2010 (2) SA 415 (CC); 2010 (3) BCLR 177 (CC) at para 70. See also More v Minister of Co-operation and Development 67 THERON J [169] In addition, as the Labour Court correctly held, the protection which section 66(2)(c) affords secondary employers is denuded if that section is interpreted to permit the aggregation of secondary employers for purposes of the proportionality inquiry. By way of example, suppose that there were two secondary strikes in support of the same primary strike and that one of these secondary strikes had a profound effect on the business of the primary employer, but that the other had a very slight effect and had serious adverse effects on the secondary employer. On the interpretation adopted by the first judgment, the latter strike might well be permissible under section 66(2)(c) even though, by hypothesis, it has only a very slight effect on the primary employer and very serious adverse effects on the secondary employer. In other words, a strike which, on its own, would be unlawful in terms of section 66(2)(c), is transmogrified into a lawful strike because of secondary strikes at entirely unconnected places of employment. That cannot be. As a result, and accepting as I do that section 66(2) is intended to afford employers a measure of protection, I cannot agree that section 66(2)(c) allows for an aggregation of secondary strikes for purposes of the proportionality assessment. [170] Save for this qualification, I concur in the first judgment. 1986 (1) SA 102 (A) at 115B-D and Minister of the Interior v Machadodorp Investments (Pty) Ltd 1957 (2) SA 395 (A) at 404D-E. 68 For the Applicants: For the Second Respondent: For the Third, Fourth, Sixth, Seventh, Eighth and Tenth Respondents: P Kennedy SC and F Boda SC instructed by LDA Attorneys Incorporated A Snider SC and Y Ntloko instructed by Cliffe Dekker Hofmeyr Incorporated A Myburgh SC and R Itzkin instructed by Webber Wentzel and ENS Africa Incorporated