A Tale of Two Citadels: Constitutional Court Resilience Against Creeping Autocratisation in India and South Africa

1 Introduction

India and South Africa used to be thought of as two of the world’s most successful post-colonial democracies. India, for its part, was regarded as having confounded mainstream political science predictions about the viability of that form of government in a large, ethnically diverse and religiously divided society. South Africa’s transition, too, was celebrated as a miracle — a rare example of a negotiated transfer of political power to a national liberation movement with a commitment to constitutionally-regulated social change. Today, however, neither country is thought of as being particularly exceptional. India under the Bharatiya Janata Party (BJP) has slid down the democracy rankings, while South Africa remains a highly unequal society in which corruption undermines efforts to address poverty and fuels populist opposition to the 1996 Constitution. While the trajectories of decline in each country are different, both countries are facing significant threats to their democratic orders.

Given this shared predicament, it is significant that South Africa’s courts have performed relatively better than India’s, with its Constitutional Court (“CCSA”) in particular still regarded as an important bulwark against de-democratisation. This was not a foregone conclusion. After populist President Jacob Zuma’s ascension to office in 2009, many feared that the governing African National Congress (ANC) would use its sustained electoral majority to subordinate the Constitutional Court.^[1] These fears were heightened when President Zuma appointed one of the Court’s most junior judges, Mogoeng Mogoeng, to the position of Chief Justice.^[2] As things turned out, however, the Court performed relatively well under Mogoeng CJ’s leadership. In one particularly noteworthy decision, it ordered the repayment of public funds that had been used to extend President Zuma’s private homestead.^[3] The decision eventually led to President’s Zuma’s departure from office and the return of a more moderate, constitution-respecting faction of the ANC to power.^[4]

The story in India has been very different. Immediately after its election victory in 2014, the BJP showed its hand by introducing a constitutional amendment to the so-called “collegium system” for judicial appointments.^[5] The Supreme Court initially resisted this move, striking the amendment down as an unwarranted political intrusion on judicial independence and thus unconstitutional under its famous “basic structure” doctrine.^[6] The story since, then, however, has been one of steady capitulation to the BJP’s ethno-nationalist agenda. Successive chief justices have used their control of the Court’s docket to delay deciding important constitutional cases or to appoint judicial panels favourable to the government.^[7] The two-year term of Chief Justice DY Chandrachud from 2022–2024 was seen by many as an opportunity for the Supreme Court to more forcefully resist India’s democratic decline, but these hopes have been dashed. The BJP, it appears, has been able to get around the failure of its 2014 constitutional amendment by using various informal measures to control India’s senior judiciary.^[8]

The purpose of this paper is to explore the reasons for the contrasting performances of the CCSA and the Indian Supreme Court and, in so doing, contribute to understanding the factors that condition the capacity of apex courts with constitutional jurisdiction (hereafter just “constitutional courts”) to resist creeping autocratisation.^[9] The next section starts by surveying the literature on this topic. It proposes that the factors conditioning constitutional court resilience against democratic retrogression may be classified into three main categories: structural, institutional and agentic. Section 3 addresses questions of research design and methodology, justifying the choice of India and South Africa as sites of comparison and explaining the approach adopted in this study. Sections 4 and 5 assess the Indian and South African cases in turn, starting with a survey of the three sets of factors identified in Section 2, and then moving on to the performance of their respective apex courts. Section 6 analyses possible explanations for the different outcomes in the two countries and ascribes the stronger performance of the CCSA to a combination of structural, institutional and agentic factors. Section 7 concludes.

2 Factors Impacting Constitutional Court Resilience

The literature on the role of constitutional courts in promoting democracy typically focuses on their capacity for “independence” or “assertiveness”.^[10] With a little adjustment, however, much of this literature has implications for the theme of this special issue — constitutional court resilience. What “resilience” adds to the usual categories, it is here assumed, is a concern for constitutional courts’ capacity to continue playing their constitutionally appointed role in the face of anti-democratic attack. It is one thing, in other words, to demonstrate independence in the context of a reasonably well-functioning democracy, and another in conditions of autocratisation.

In the brief discussion that follows, the various factors that the judicial independence literature has identified as conditioning the independence of courts are classified under three headings — structural, institutional and agentic — and then repurposed to fit the focus of this issue. The three headings are purely classificatory. They are intended to capture the three main determinants of social processes typically emphasised in the social sciences. No argument is made about their relative importance in explaining constitutional court resilience or how they interact with each other in different contexts.

By “structural” is understood factors that have to do either with party-political dynamics or with deep-seated cultural attitudes to law and legal institutions. As to the first, the literature on judicial independence has identified “political fragmentation” as a particularly significant variable, with judges’ ability to play an effective role in politics having been found to strengthen as power is dispersed away from a single political faction.^[11] There is no reason to think that this finding would not also hold for constitutional court resilience. If resilience is understood as a capacity to withstand anti-democratic attack, then we should expect that courts would be better able to display that quality the more dispersed political power was in the society concerned.

The second structural factor concerns deep-seated cultural attitudes to law and legal institutions. In the judicial independence literature, it has thus been suggested that independence is supported by the existence of a relatively sticky cluster of social norms that raises the costs to political actors who seek to attack the courts.^[12] These norms may strengthen over time, such that past events are capable of bequeathing a “legacy” of respect for judicial independence, even in the face of major institutional changes and shifts in the concentration of political power.^[13] Again, this factor ought to apply equally to the question of constitutional courts’ ability to withstand anti-democratic attack.

By “institutional” is meant formal constitutional-design features that affect how judges are appointed and the degree of protection they enjoy from attack. There is a wide range of well-known factors of this sort that have been said to condition judicial independence and, by extension, also conceivably constitutional court resilience. These include such factors as security of tenure and remuneration,^[14] the jurisdiction of the court (e.g. how wide it is, whether the court’s constitutional review mandate is express or not, and whether it extends to both ex ante and ex post review),^[15] the presence of “fourth-branch” institutions supporting constitutional democracy,^[16] and the judicial appointments process, including the extent to which that process is politicised, not just in the sense that it allows for the involvement of political actors, but also in the sense that the known political allegiance of the candidates for judicial office matters in whether they are appointed or not.^[17]

In respect of the latter issue, judicial appointment models may be distinguished according to how they balance two competing imperatives: the need, on the one hand, to provide for some political involvement in the appointments process to ensure that judges are held accountable for their decisions and the need, on the other, to ensure that political involvement not go so far as to undermine independence.^[18] The preferred modern approach is to use some combination of executive appointment and scrutiny of candidates for office by a non-partisan judicial service commission.^[19] The performance of purely political models as against models that provide for some cross-party or civil society involvement has not been systematically tested.

The final set of factors that is thought to condition judicial independence is “agentic” in the sense that they relate to “the choices justices make”, both as individuals and groups.^[20] This set of factors has its origin in behaviouralist studies of judicial decision-making in the United States,^[21] but it has since expanded to include rational-choice studies of judges as institutionally-constrained actors whose first-preference choices may be impacted by their immediate institutional setting and/or the broader political environment. By “choices” in this context is generally meant decision-making choices (both as to doctrine and as to policy outcome).^[22] But this set of factors also extends to other discretionary choices, such as the way constitutional courts choose to engage their public audience through their media strategy, and their informal relationship to civil society groupings and social networks (e.g. the organised legal profession).^[23]

Of this broad set of agentic factors, decision-making factors have understandably received the most attention. The last 20 years or so have thus seen the three models of judicial behaviour originally developed to explain decision-making on the US Supreme Court (legal, attitudinal and strategic) extended to constitutional courts in other settings, including autocratic settings or “fragile democracies”.^[24] The two main theorisations that have emerged from this comparative work are a developed version of the strategic model, which conceives of justices as political actors rationally calculating how best to ensure that their decisions are enforced,^[25] and an “adjudicative-strategy” model that stresses the way in which judges who operate in legalist constitutional cultures, where law constrains political decision-making to some degree, must work in and through the law to maximise their capacity to speak legal truth to political power.^[26]

3 Research Design

India and South Africa are appropriate choices for examining the factors conditioning constitutional court resilience against autocratisation for two main reasons. First, they are both post-colonial democracies whose judiciaries have at times been celebrated for their independence, and yet both are also countries that, over the last 15 years or so, have experienced increasing levels of autocratisation. Second, in the face of this democratic retrogression, South Africa’s Constitutional Court is generally seen as having performed better than the Indian Supreme Court. This combination of similarity in past performance and difference in contemporary outcome allows us to isolate the factors that may be responsible for this result.

As to their past record, both the Indian Supreme Court and the South African Constitutional Court were once celebrated for their assertiveness. The Indian Supreme Court’s heyday was in the 1980s, when its public interest litigation (PIL) jurisprudence attracted global attention.^[27] Similarly, the CCSA was for a time the toast of the comparative constitutional law community, with its decisions celebrated in the 1990s and early 2000s for their doctrinal creativeness and wise judicial statesmanship.^[28] Neither court’s record, to be sure, was a certain predictor of future resilience against autocratisation since both courts had built their reputations during periods of political support for independent judicial review. In India, the two judges who drove PIL in the 1980s (Bhagwati and Krishna Iyer JJ) were initially operating in an environment in which Prime Minister Indira Gandhi had explicitly stated her preference for “committed” judges.^[29] In the case of South Africa, the CCSA famously had the support of President Mandela as it began its work.^[30] Both courts could thus be said to have been operating under propitious conditions. Nevertheless, the fact that they exploited these conditions to assert their role means that, in theory at least, they had built some scaffolding for “resilience” during tougher times of autocratisation.

At the same time, India and South Africa also stand for something distinctive in comparative constitutional studies. They are both countries that are thought to have had relatively robust constitution-making processes during which a sincere attempt was made to adapt liberal constitutionalism to the challenges of post-colonial governance.^[31] In both countries, too, the idea of transformative constitutionalism (TC) has been used to describe the distinctive form of constitutionalism that emerged. In contrast to classic liberal constitutionalism, TC connotes a form of constitutionalism in which state institutions are under a duty to drive the transformation of society towards a constitutionally-imagined conception of post-colonial justice.^[32] Constitutional courts under this form of constitutionalism are given an explicit mandate to ensure that state actors stick to the constitution’s social transformation agenda. That role should not be exaggerated — judges under TC are meant to be facilitators not dictators-but it is nevertheless relevant to the question of resilience that in both India and South Africa there is an expectation that their apex courts will protect the democratic order against attack.

In both India and South Africa the shine has recently come off the constitutionally-mandated social transformation project. While the underlying causes are different, the last 15 years have seen a rising decolonial critique of the original constitutional settlement in both countries.^[33] Two constitutions that were once thought to be prime examples of the capacity of liberal constitutionalism to transcend its origins in the European Enlightenment are now charged with having been drafted by Westernised elites who reproduced the culturally-alienating conceptual frameworks of their former oppressors. In both countries, too, the decolonial critique — once the preserve of public intellectuals and academics — has recently been taken up in party politics. In India, the BJP has recently used decolonial arguments to justify a series of criminal law amendments.^[34] In South Africa, the 2024 national elections saw the emergence of a political party, the uMkhonto we Sizwe Party (MKP), that has explicitly committed itself to abolishing the 1996 South African Constitution and replacing it with a more authentic, Africanist document.^[35] In both countries, in other words, autocratisation is taking the form of on an attempted ethno-nationalist reworking of the original, more inclusive constitutional settlement. In the course of this, the Supreme Court of India and the CCSA have been targeted as symbols of their society’s allegedly pro-Western founding order.

In so far as it is measurable, the degree of autocratisation that has occurred over the last 15 years in India and South Africa is roughly comparable. Freedom House currently ranks India as “partly free” with a combined score of 63 (with scores of 31/40 on political rights and 32/60 on civil liberties).^[36] South Africa, for its part, is ranked as “free” with a score of 81 (34/40 and 47/60 respectively). The variance of from 63 to 81 might seem significant but in fact the score in respect of political rights is similar. India is also only just inside the “partly free” classification on the Freedom House Index, needing just three more points on the civil liberties scale to be classified “free”.

The Economic Intelligence Unit’s 2024 Democracy Index: What’s Wrong with Representative Democracy lists both India and South Africa as “flawed” democracies, with India on a score of 7.29 (ranked 41) and South Africa on 7.16 (ranked 43). Thus, on this measure, the two countries are very close to each other. Historical figures for 2006–2023 show India dropping from 7.68 to 7.18 with an even more dramatic dip between 2019 and 2021. South Africa dropped from 7.92 to 7.05 over the same period. On this scale, then, democratic decline for South Africa has been a little more severe than in India.^[37] Together, the two sets of indicators suggest that the level of autocratisation in India and South Africa is close enough as to make comparison possible.

4 The Indian Supreme Court’s Resilience Against Autocratisation Under the BJP

2014 marked a significant shift in India’s democracy, with the BJP winning an absolute majority in the lower house, the Lok Sabha — the first time a single political party had been able to do that in 30 years.^[38] Since, 2014, the BJP has continued to dominate Indian politics, winning an even larger majority in 2019,^[39] before slipping back slightly in 2024.^[40]

It is not just the BJP’s electoral dominance that threatens India’s democracy but also the particular nature of the challenge that it poses to the 1946–49 constitutional settlement. The latest in a long line of Hindu nationalist parties, the BJP is expressly committed to transforming India into a Hindu Rashtra (a religio-cultural state based on Hindu values).^[41] Its ideology of Hindutva represents a crude, ethno-nationalist understanding of Hinduism that runs counter to its best interpretation and the Indian sub-continent’s tradition of religious toleration more generally.^[42] From 2014, the BJP has successfully been able to mobilise anti-Muslim sentiment in India for political gain. In so doing, it has wrested political power from the formerly dominant Congress party, which is identified with the Constitution’s founding commitment to secularism. The BJP has also consolidated India’s powerful security state, which predates its rise but which the BJP has been able to strengthen in the era of alleged Muslim terrorism.^[43]

While the BJP has not yet sought to abolish India’s 1950 Constitution, it has introduced a number of measures that have been described as killing off constitutionalism “with a thousand cuts”.^[44] The most significant of these have been the abrogation of Article 370 of the Constitution, which had preserved the special status of the disputed territory of Jammu and Kasmir within an asymmetrical federal system,^[45] and the adoption of the Citizenship Amendment Act, which provides for differential access to citizenship for asylum seekers according to religion.^[46]

As noted in the Introduction, the Indian Supreme Court’s performance in the face of these anti-democratic attacks has declined over time. Ten years ago, the Supreme Court was still thought to be “one of the most powerful constitutional courts in the world”,^[47] mainly because of its control over the judicial appointments process, which was extended over a series of Judges’ Cases.^[48] In terms of the so-called “collegium system”, the Chief Justice and several of the most senior members of the Supreme Court had effective control over appointments to the High Courts and the Supreme Court itself.^[49] Shortly after it swept to power in 2014, however, the BJP introduced a constitutional amendment to change this system and instead provide for the appointment of judges by a National Judicial Appointments Commission (NJAC).^[50] In addition to the Chief Justice and two other senior Supreme Court judges, the proposed NJAC gave a role to the Union Minister of Law and Justice and two civil society members. While the collegium system was flawed in many ways, this proposed change was seen by many as an attempt to exert greater political control over the judiciary. In the event, the Supreme Court struck the amendment down in a unanimous judgment. Judicial independence, the Court held, was integral to the functioning of India’s democracy and thus the attempt to dilute senior Supreme Court judges’ influence over judicial appointments went against the Indian Constitution’s “basic structure”.^[51]

The Supreme Court’s striking down of the amendment momentarily appeared to confirm its status as the world’s most powerful constitutional court. As it turned out, however, the BJP was thereafter able to use other, more informal measures to take control of the higher judiciary. Under the collegium system, the Chief Justice is required to consult the Law Minister in regard to judicial appointments. It now appears that the BJP has been able to use this aspect of the process to delay appointments of capable candidates and to insist on the appointment of BJP loyalists.^[52] There is also credible evidence that judges have ruled in favour of the government in return for lucrative appointments after retirement from the Bench.^[53] Over and above these two mechanisms, the BJP has also been able to exert ideological control over the judiciary through its ability to influence the nature and composition of India’s “power elite”.^[54] Since the higher judiciary is ultimately drawn from that elite, this has gradually ensured that judges sympathetic to the BJP’s political programme have come to dominate the upper echelons of the judiciary.^[55]

In addition to the judicial appointments process, other institutional factors potentially affecting the resilience of the Supreme Court include the Chief Justice’s power to establish so-called “Constitution Benches”, which is loosely regulated and thus susceptible to deployment either in defence of democracy or to pack the Bench with party loyalists.^[56] The Chief Justice is also the Master of the Roster, with the power to choose which judges hear politically sensitive cases.^[57] Finally, constitutional cases comprise only a small proportion of the Supreme Court’s immense docket, meaning that the Chief Justice is able to rely on the busyness of the Court’s regular case load to justify delays in hearing important constitutional cases.^[58]

While these institutional features could in theory be used for purposes of resilience, the secondary literature on the Supreme Court’s performance now conclusively shows that they have been used to avoid principled confrontations with the BJP on sensitive political issues. Thus, for example, the hearing of the so-called “Electoral Bonds Case”, which concerned regulation of party-political funding, was delayed for many years by successive chief justices before finally being decided by the Chandrachud Court. By the time it was decided, the BJP had already drawn maximum benefit from what was eventually declared to have been an unconstitutional state of affairs.^[59] Similarly, in the disputed union territory of Jammu and Kashmir, the Supreme Court has not heard any of the numerous habeas corpus cases that have been filed.^[60] Instead, it has allowed the use of “sealed covers” to ostensibly preserve sensitive security information but in practice to sustain an information blockade over security-police operations.^[61]

The large number of justices on the Supreme Court and unregulated procedures for assignment of judges to benches have led to a further series of problems. The first concerns the phenomenon of the “polyvocal court” in terms of which different Benches give sometimes diametrically opposed interpretations of the law.^[62] In practice this means that apparent doctrinal advances in defence of democracy can quickly be undone by contrary decisions. The second problem is that, unlike the South African Constitutional Court, the Indian Supreme Court — even when intervening in defence of democracy — has tended to deal with democratic dysfunction by taking over those duties itself rather than shoring up other institutions.^[63]

Beyond provisions directly touching on the judiciary, one other institutional issue concerns whether the Indian Constitution really provides all the protections for judicial resilience one might expect. Because of the Supreme Court’s reputation during the period of PIL, the Indian Constitution is often thought of as a progressive, pro-poor Constitution – if not by design, then by dint of the way the Court has been able to interpret it.^[64] Recent revisionist accounts of the Supreme Court, however, push back against this picture. For example, Anuj Bhuwania has argued that PIL started to morph from the 1990s into a device for defending the judiciary’s own middle-class interests, in a way that demonstrates the alignment of the judiciary with India’s political elite and which potentially undermines its capacity to resist the BJP’s hegemony.^[65] Newer work by Bhuwania reminds us that the Indian Constitution was never very strong on civil rights, and that the Supreme Court has always been weak in defending individual liberty against the security state, particularly in cases concerning the Unlawful Activities (Prevention) Act (UAPA).^[66]

When Chief Justice Dhananjaya Yeshwant Chandrachud was appointed to the Court in 2022, it was widely hoped that he would be able to arrest the poor performance of the Supreme Court to that date. Chandrachud had, by the time of his appointment, established a reputation as one of the Supreme Court’s most liberal and hard-working justices. His two-year appointment was also set to be longer than the norm, giving him scope to make a difference.^[67] Unfortunately, these expectations proved to be unfounded. While there is some evidence that Chandrachud CJ took a more rigorous approach to his function as Master of the Roster, and while some lingering constitutional issues were resolved, the emerging consensus among commentators shortly after the end of his term is that he was not as effective as he might have been, both in terms of his personal conduct and in terms of his decision-making record.

As to the former, two extra-curial actions by Chandrachud CJ attracted tremendous censure. The first was his decision, not only to perform an “aarti” (a type of religious ceremony) for Ganesh Puja with Prime Minister Modi in his home, but also to invite television cameras to film the ceremony.^[68] The second concerned a speech Chandrachud delivered in his home town during which he remarked that he had asked for divine inspiration in the Ayodhya Case (discussed below). The remarks not only suggested that Chandrachud might have allowed extra-curial considerations to sway his views. They were also suggestive of a possible conflict of interest.^[69]

In addition to these extra-curial actions, two decisions — one as a puisne justice and the other as Chief Justice — call into question Chandrachud CJ’s capacity to speak legal truth to political power.

The first concerned the long-running dispute over a parcel of land at Ayodhya, which Hindus claimed as the birthplace of Ram but which, until 1992, had been the site of a mosque (the Babri Masjid) built by the Mughal emperor Babur.^[70] “Had been” because, in that year, Hindu religious zealots razed it to the ground. The demolition of the mosque and its replacement by a Hindu temple had long been a demand of Hindu nationalist groups, going all the way back to the nineteenth century.^[71] After 1992, the fate of the disputed land at Ayodhya became central to the mobilisation of anti-Muslim sentiment that eventually carried the BJP to power in 2014. Five years later, the dispute was eventually settled by Supreme Court, when Chandrachud was an ordinary justice. In a lengthy unanimous judgment that many believe he authored, the Court awarded the land to the Union Government in trust for the building of a temple.^[72] Prime Minister Modi duly celebrated the construction of the Ram Mandir in 2024, just in time for the national elections held that year.^[73]

The second case concerned Jammu and Kashmir, which had been accorded a special status by Article 370 of the Constitution within India’s asymmetric federal system.^[74] After its second election victory in 2019, the BJP moved to give effect to another longstanding part of its political programme by abrogating Article 370 using a highly questionable combination of constitutional amendments and executive acts.^[75] In 2024, Chandrachud CJ, nearing the end of his tenure, authored the Court’s decision upholding the constitutionality of these measures, again deciding in favour of the BJP government on grounds that many found unpersuasive.^[76]

In summary, the Indian Supreme Court’s resilience against democratic retrogression under the BJP has been marked by steady decline since 2014.

5 The CCSA’s Resilience Against Autocratisation Under President Jacob Zuma

The CCSA has operated for most of its life in a one-party dominant democracy, with the African National Congress (ANC) consistently winning more than 60% of the national vote from 1994–2019. According to the political fragmentation thesis, that means that the CCSA should have been a constrained court for much of this time, its capacity independently to enforce the Constitution limited by the ANC’s ability to attack it in various ways, including by stacking the Bench and through the non-implementation of its decisions. In fact, things did not turn out this way. In the first 10 years of its life, at least, the CCSA built an international reputation for innovative decisions that creatively gave effect to South Africa’s state-of-the-art, “post-liberal” Constitution.^[77] Domestically, the Court enjoyed strong public support and was able to insert itself as a powerful veto player in South Africa’s nascent constitutional democracy.^[78] While international interest in the Court has declined after the departure of the original Bench, the CCSA has continued to play a forceful role in South African politics, as this section will briefly illustrate.

The CCSA’s strong start and enduring institutional power partly had to do with President Mandela’s steadfast support for the Court and its role in nation-building reconciliation.^[79] It also had something to do with the calibre of the justices on the original Court and the considered way they approached the implementation of their mandate.^[80] As a structural matter, however, the Court’s capacity to assert its independence also undoubtedly had something to do with South Africa’s rule-of-law tradition. Even under apartheid, the courts were not entirely dependent. Operating as they did in a political context defined by the ruling National Party’s claimed attachment to Western values, including the rule of law, liberal judges were able to exploit gaps in the legal system to hold the government to account.^[81] This record of independent adjudication under apartheid functioned as a cultural inheritance of sorts, which both informed the drafting of the 1996 Constitution and fed into the public support that the Court was able to build.^[82]

For the first 15 years of its existence, South Africa’s rule-of-law tradition served the CCSA and, in turn, South African democracy well. The Court was also undoubtedly assisted by the fact that President Nelson Mandela strongly supported its independence. From about 2009, however, new challenges presented themselves. In that year, Jacob Zuma was able to depose Thabo Mbeki as President. There followed a period of populist political leadership, mismanagement of public finances and widespread corruption, which has come to be referred to as the era of “state capture”.^[83] While not amounting to out-and-out autocratisation, this period was characterised by a sustained assault on independent constitutional institutions.^[84] Elections continued to be held and were generally regarded as free and fair. Civil and political liberties, including free speech, were also respected. But there was at the same time a systematic attempt to collapse the distinction between the ANC as the governing political party and the state. Through a policy known as cadre deployment, the ANC staffed formally independent state institutions with party loyalists. Various public corporations, such as the state electricity provider, Eskom, and the rail corporation, Transnet, were captured and bent towards corrupt purposes. Particularly severe was the assault on the National Prosecuting Authority as Zuma sought to avoid prosecution for his involvement in the so-called “Arms Deal”.^[85] The CCSA itself was attacked in various ways: through the non-appointment to senior leadership positions of judges who were thought to be unsympathetic to President Zuma,^[86] by the launching of a dubious investigation into its record in promoting socio-economic rights,^[87] and in the form of public attacks by prominent ANC politicians, including Lindiwe Sisulu, the national Minister of Transport and daughter of former ANC President, Walter Sisulu.^[88] As noted in Section 3, South Africa’s position on the Freedom House and Economic Intelligence Unit indices began to slide during this time.

The 1996 Constitution remained substantially unchanged during this time, with all of the institutional features that had been so celebrated still nominally intact. These included strong textual support for the Court’s counter-majoritarian function, and the presence of a range of so-called “Chapter 9 institutions” protecting constitutional democracy, including a Public Protector and Judicial Service Commission (JSC).^[89] While the composition of the JSC theoretically allows a dominant political party to appoint the majority of its members,^[90] the Commission in fact operated relatively independently during the first 15 years of the CCSA’s life. This meant that, at the start of the State Capture period, when Mogoeng Mogoeng was nominated for the position of Chief Justice, the JSC was able to hold fairly robust civil society hearings that interrogated his past record as a High Court judge.^[91] These hearings arguably proved to be significant later, when Mogoeng Mogoeng was called on to enforce the Constitution against the Zuma faction of the ANC.^[92] Having been so thoroughly interrogated by the JSC, he was evidently determined to demonstrate his independence.^[93]

At the same time, the CCSA was undoubtedly buoyed up during the troubled 2009–2019 period by the strong start that the Chaskalson, Langa and Ngcobo Courts had made.^[94] During the first 15 years of its institutional life, the CCSA had developed a number of adjudicative strategies that enabled it to enforce the Constitution without triggering a debilitating attack. Among these were the CCSA’s ability to develop flexible doctrines that allowed it to adjust its decisions in controversial cases to the “micro-politics” of the case, while still maintaining a reputation for decision-making according to law.^[95] Importantly, the CCSA did not openly declare the politicality of its function, as a very influential early paper authored by American CLS scholar, Karl Klare, had urged it to do.^[96] Rather, the CCSA adopted a classic legalist posture, justifying its authority to strike down legislation as a function of the apolitical, technical nature of its mandate.^[97] The other very significant thing that the CCSA did was never to seek to usurp the political branches’ functions by, for example, imposing intrusive remedial orders that required it to play a hands-on role in governance.^[98] Rather, the CCSA during the first 15 years of its existence sought to shore up the capacity of the political branches to perform their role in the constitutional system. Unlike the Indian Supreme Court’s usurpation of the role of democratic institutions during the heyday of its PIL jurisprudence, this was an approach that promised to stand the Court in good stead during times of democratic retrogression.

In 2009, when Mogoeng was appointed as Chief Justice, many feared that the CCSA would not be up to the task of resisting the assault on democratic institutions that the Zuma Presidency was clearly planning. Mogoeng, it was thought, was a junior justice who had been appointed largely because he shared Zuma’s socially conservative views and would be politically pliable. As things turned out, however, Mogoeng was a fiercely independent Chief Justice. During his tenure from 2009–2019, the CCSA handed down a number of forceful decisions. Space allows only for a brief commentary on a few of them to give a flavour of the resilience the CCSA displayed. Particular attention is paid to cases concerning the defence of constitutional institutions in the face of President Zuma’s state capture project. Since that was the particular form of autocratisation that South Africa was going through, it makes sense to focus on those.^[99]

The most significant of these cases was undoubtedly the CCSA’s judgment in the Nkandla Case.^[100] The backdrop was quite convoluted, but in essence it involved an allegation, first made in a 2009 newspaper report, that public funds had been used for security upgrades to President Zuma’s private residence. The upgrades included the building of a swimming pool, a cattle kraal, a visitors’ centre, an amphitheatre and a chicken run among other things.^[101] Following the report, the Public Protector’s office began investigating the lawfulness of the expenditure. It issued its findings in March 2014.^[102] Finding that the President had breached his legal obligations, the Public Protector directed him to pay back the personal portion of the expenditure. When the President failed to comply with this direction, two minority political parties, the EFF and the DA, launched legal proceedings to enforce the remedial action the Public Protector had required. The two central issues for determination were whether the remedial action had created binding legal obligations and whether the President and the National Assembly had violated their constitutional obligations.

The CCSA’s decision, authored by Mogoeng CJ, began by depicting constitutionalism as “the sharp and mighty sword that stands ready to chop the ugly head of impunity off its stiffened neck.”^[103] “One of the crucial elements of [South Africa’s] constitutional vision”, Mogoeng CJ continued, “is to make a decisive break from the unchecked abuse of State power and resources that was virtually institutionalised during the apartheid era.”^[104] By tying the CCSA’s role in protecting democracy to the country’s apartheid history in this way, Mogoeng CJ was able to present the 1996 Constitution as a crucial bulwark against democratic retrogression. Not just that, but the CCSA’s decision focused on shoring up the role the Public Protector had already played in bringing the misuse of taxpayer money to public attention. The major doctrinal innovation in the case was thus the CCSA’s decision that the Public Protector’s remedial actions should be understood as legally binding.^[105]

In a second significant case, Glenister v. President of the Republic of South Africa and Others,^[106] a narrow majority of the CCSA struck down parts of a statute regulating the establishment of a new anti-corruption agency (“the Hawks”). The case followed a controversial decision to close down an earlier, very effective agency (“the Scorpions”).^[107] The decision to close down the Scorpions and replace it with the Hawks had been taken at an ANC national conference following a humiliating investigation by the Scorpions into ANC parliamentarians’ abuse of travel allowances. While the Scorpions had been based in the formally independent National Prosecuting Authority, the Hawks were part of the South African Police Services, and entirely under the control of the relevant minister. Significantly, in its decision, the CCSA did not order the resurrection of the Scorpions. Rather, it struck down just those parts of the challenged statute that undermined the Hawks’ independence.^[108] In so doing, the CCSA developed a series of criteria guiding the legislative creation of independent state institutions, which have been applied in subsequent cases.^[109]

As this brief overview shows, the CCSA’s approach in cases involving threats to the democratic order is centred around shoring up the functioning of independent state institutions rather than taking them over. In this way, it has been able both to respect a somewhat traditional conception of the separation of powers while nesting itself within, and reinforcing, a set of institutions (including new “fourth-branch” institutions) that together work cooperatively to protect the democratic system from attack.^[110]

6 Explaining the Divergent Outcomes

In what is admittedly a very cursory sketch, it is apparent that the CCSA, amidst comparable levels of autocratisation, has fared better than the Indian Supreme Court. This difference, to be sure, might be attributable in part to variation in the nature of the autocratisation process in each case. In the Indian case, the Supreme Court faced a coherent political party that had been building the ideological platform for its challenge to India’s secular constitutional settlement for decades. Through its association with the Rashtriya Swayamsevak Sangh (RSS), the BJP had built strong social networks that it was able to use to reach deep into democratic institutions, populating them with ideological loyalists. In South Africa, by contrast, President Zuma’s challenge to the democratic order was more opportunistic, driven as it was by the desire for private financial gain rather than a coherent ideology as such. Being a faction within the ANC, Zuma loyalists were also confronted by internal opposition within the party still loyal to the constitutional settlement. Even so, the CCSA performed remarkably well, and thus it is worth exploring whether there were relevant differences in the three sets of factors conditioning constitutional court resilience, as set out in Section 2.

As to structural factors, we should note first the major difference that, in the Indian case, during the time period of interest, politics had begun to be less fragmented than it had been for some 20 years.^[111] After two decades of coalition government, the BJP in 2014 won an outright majority in the Lok Sabha. It then increased that majority in 2019. If the political fragmentation thesis holds, that would suggest an increased capacity on its part to subordinate the courts. Not just that, but the BJP was not the party of national liberation, but rather the heir to parties that had played only a muted role in the 1946–1949 Constituent Assembly debates. While this still did not mean that the BJP could directly attack the constitutional settlement, and demonise the Court as an agent of that settlement, it meant that it did have less loyalty to the politically inclusive principles that had originally informed it. For the BJP, driven as it was by its understanding of India as a civilisational state, moving against the judiciary was part of a long-term project.

South Africa for the time period in question, 2009–2019, was also a dominant-party democracy, with the ANC winning successive elections. And yet the CCSA was apparently better able to assert its independence. One part of the reason for that might have been that the ANC was internally divided, meaning that there was political fragmentation of sorts, just not at the party-political level. It is certainly the case that the competing factions within the ANC, as well minority political parties, looked to the CCSA to resolve internecine fights. If so, this need for the Court as a neutral umpire might have replicated some of the conditions for judicial independence that the political fragmentation thesis picks up. And yet, the Zuma faction was able to undermine other state institutions, including the National Prosecuting Authority and, after a brief show of defiance in the Nkandla Case, the Public Protector.^[112] The fact that the ANC was internally divided, therefore, cannot be the sole explanation.

We must thus complement the political fragmentation thesis by having regard to another structural factor — deep-seated cultural attitudes to law and legal institutions. Here, a very big difference emerges between India and South Africa. Both societies have strong legal complexes, and highly visible and prestigious legal professions. But if we look to the history of the liberation struggle, the South African struggle was more significantly waged through law than the Indian. From the 1920s, when Gandhi took control of the Congress party, the main devices used to wage the liberation struggle were mass action and defiance. South Africa’s liberation struggle, by contrast, included a significant strategic litigation component.^[113] That history bequeathed to South Africa a strong tradition of judicial independence, including especially in relation to matters of personal security.^[114]

This legal-cultural difference must be read alongside significant institutional differences as well. As the more modern of the two constitutions, the 1996 South African Constitution was able to draw on a richer tradition of constitutionalist experimentation than the Indian. This is reflected, for example, in its more comprehensive rights charter, including strong administrative-law rights, rights to freedom of information and expression, and protections for journalists.^[115] The 1996 Constitution also contained, as we have seen, much more explicit provision for fourth-branch institutions. While the Public Protector and the JSC were eventually attacked by the Zuma faction, they were able to play crucial roles at critical stages in keeping the channels of democratic participation open and supporting the groundswell of opposition that eventually led to Zuma’s ousting.^[116]

One crucial but potentially misleading institutional factor concerns the judicial appointments process. While the Indian collegium system nominally looks more supportive of judicial independence than South Africa’s JSC, it proved relatively easy to suborn through informal pressure once the BJP’s attempt to amend it constitutionally failed. South Africa’s JSC, by contrast, looks formally like it allows for political influence, but in fact the public nature of its operation makes it much harder to exert informal pressure on judicial appointments. Even where this occurs, the public nature of the process can expose its abuse. Not just that, but Justices like Mogoeng Mogoeng, who was appointed under a cloud, may have an added reason to prove their detractors wrong. This suggests that formalising the judicial appointments process, and particularly providing a role for civil society, may promote greater resilience than leaving the process in the hands of judges. While the Supreme Court’s striking down of the NJAC Amendment was understandable in the circumstances it occurred, it does not in the end appear to have insulated the courts from political pressure.

Agentic factors, the third set, are also quite significant in explaining the CCSA’s superior resilience. If we compare the heydays of the two courts, the Indian Supreme Court’s PIL phase was built around creative interpretations of the Constitution that sometimes flew in the face of the constitutional text.^[117] In its pursuit of a pro-poor reading of the Constitution and broadening access, the Indian Supreme Court threw many traditional aspects of the separation of powers out of the window, including rules of access to the court and remedial powers. While these innovations were celebrated at the time, their capacity to work as a force for good was highly contingent on evanescent political circumstances — Indira Gandhi’s support for the Court and judges committed to pro-poor adjudication. When these contingent factors disappeared, the doctrines the Court had created proved ill-equipped to enable the Court to deal with the BJP. On the contrary, the Court’s PIL jurisprudence, morphing as it did into middle-class jurisprudence, in fact made it more vulnerable to influence by the new power elite. The Court’s doctrinal blurring of the usual separation of powers meant that the divide between it as the enforcer, and the political branches as the implementer, of the constitutional project, broke down. The Court, it turned out, had spent too long taking over governmental functions and too little time shoring up democratic institutions. When combined with the Supreme Court’s huge docket, the small proportion of constitutional cases in that docket, the Chief Justice’s power to manipulate the outcome of decisions through the Master of the Roster function and the power to appoint Constitution Benches, the Supreme Court was left with few options to counter the BJP’s growing hegemony.

The CCSA, by contrast, had the benefit of a much slower start, with a very low case load for the first 10 years of its existence, during which time it had an exclusively constitutional jurisdiction. With the support of President Mandela, the Court was able to elaborate doctrines and a general approach to its mandate that stood it in good stead for the 2009–2019 period. While the Court now has a higher case load, and plenary jurisdiction, the doctrinal platform that it laid down in the first 15 years of its existence helped it withstand the state capture era.

The CCSA did have to make one crucial shift in approach during this period. Whereas under Presidents Mandela and Mbeki, it had positioned itself as assisting the ANC in the implementation of the constitutional project,^[118] under President Zuma it was forced to shift its focus to reviewing the rationality of appointments to democratic institutions (where those opponents were of clearly unfit people whose qualification for the job was loyalty to the Zuma faction). Crucially, this approach was aimed at shoring up democratic institutions rather than taking over their functioning.

7 Conclusions

This comparison of the Indian Supreme Court’s and CCSA’s resilience in the face of autocratisation has found that all three sets of factors that are typically thought to condition that issue matter, but in ways that have not been noted before.

First, along the structural dimension, it is clear from the South African case that a deep-seated tradition of judicial independence developed during a past struggle against authoritarianism might make up for a lack of political fragmentation. Constitutional courts may further be able to stand up to a dominant party more easily when that party is the party responsible for adopting the Constitution rather than an insurgent party whose political programme is in many ways opposed to the original constitutional settlement. This explains why the political fragmentation thesis works as an explanation for the Indian Supreme Court’s comparative lack of resilience between 2015 and 2019 but not for the CCSA’s record between 2009 and 2019.

Second, along the institutional dimension, a nominally court-controlled judicial appointments process may be easier to influence than a JSC-type process, which at least forces informal pressure on the process into the open. More extensive protections for fourth-branch institutions and more extensive rights protections also appear to matter.

Finally, along the agentic dimension, a decision-making strategy that preserves the separation of powers and shores up democratic institutions appears to position the court better for resilience than one that sees it taking over governmental functions. At a more general level, constitutional courts that adopt a more legalist posture might fare better than courts that collapse the distinction between law and politics and openly pursue ideological projects through law.