Memory, Past Evils and Constitutional Justice. Lessons from the United States, Germany and South Africa

Introduction: the book in context

Justin Collings’ book Scales of Memory. Constitutional Justice and Historical Evil is about constitutional justice and the memory of past evils. More specifically, it investigates how the Supreme Court of the United States, the Federal Constitutional Court of Germany and the Constitutional Court of South Africa have grappled, respectively, with the legacies of slavery, Nazism, and apartheid. The book shows that these courts have been ‘mnemonic actors’ and ‘agents of collective memory’, ¹ and have thus contributed enormously to shaping these countries’ constitutional identities.

In general terms, Collings’ work represents an important contribution to an underdeveloped field of research, namely that of comparative constitutional history. Indeed, while the scholarly literature in comparative constitutional law is large and growing, the studies on comparative constitutional history continue to be quite scarce. ² Recently, there has been an effort to address this shortcoming, ³ but the overall impression is that comparative constitutional history still struggles to fully emerge as a distinct and independent field of study.

Within the macro-category of comparative constitutional history, this book focuses in particular on the use of history in constitutional interpretation. In the United States, the use of history as an interpretative resource in construing and applying the provisions of the Constitution is a topic that has received enormous attention from scholars. As is well known, today much of that attention centres around theories of constitutional originalism. Comparative studies of uses of history in constitutional adjudication are, by contrast, less common. While some scholars – notably Renáta Uitz, ⁴ Daphne Barak-Erez, ⁵ Ozan O. Varol, ⁶ and Jamal Greene and Yvonne Tew ⁷ – have explored how different national courts engage with historical arguments, ⁸ there remains considerable work to be done in the area. Scales of Memory, which focuses on narratives and memory in constitutional adjudication, is undoubtedly a welcome addition to this strand of research.

Collings’ book is also in conversation with scholarly works in transitional studies, and more specifically in transitional justice, which refers – according to the definition of Ruti G. Teitel – to ‘the conception of justice associated with periods of political change, characterized by legal responses to confront the wrongdoings of repressive predecessor regimes’. ⁹ Constitutional courts often played an important role in the processes of transition from autocratic regimes to democratic forms of government, ¹⁰ and in their case law they frequently (but not always, as will be shown below) invoked past evils, such as Nazism, Fascism, Communism, military regimes, colonialism or apartheid. These memories can be important legitimating resources, both for the courts themselves and for the state at large. As argued by Alec Stone Sweet, constitutional courts ‘can provide a focal point for a new rhetoric of state legitimacy, one based on respect for democratic values and rights, and on the rejection of former rhetoric (of fascism, military or one-party rule, legislative sovereignty, the cult of personality, and so on)’. ¹¹

The United States, Germany and South Africa as case studies

In Scales of Memory, Collings looks at three constitutional experiences: the United States, Germany, and South Africa. The bulk of the book revolves around the analysis of these case studies. Indeed, in addition to the introduction and the conclusion, the book consists of three parts: Part I ‘After Slavery’, Part II ‘After Auschwitz’, and Part III ‘After Apartheid’. As the author readily admits, whenever he discussed his book project with colleagues, he was asked, almost immediately, the same question: ‘Why these three cases and not others?’ ¹² Comparative scholars know very well that the selection of case studies is usually extremely challenging. The main reasons put forward by Collings to justify his choice to analyse these three case studies are convincing. In the first place, each country object of analysis ‘has a powerful and influential constitutional court that has frequently and prominently invoked an evil past’, and in each country ‘the Constitution itself is a powerful source of constitutional identity’. ¹³ Second, from an historical-temporal perspective, the constitutional texts of these countries are representative of three generations of constitutions: a first-generation constitution (the US), a second-generation constitution (Germany), and a third-generation constitution (South Africa). ¹⁴ Third, these constitutions represent, in terms of their origins, a revolution-based constitution (the US), a war-based constitution (Germany), and a pact-based constitution (South Africa). ¹⁵ Fourth, the three courts analysed in the book are also representative from a geographical standpoint: one court is from North America, one from Europe, and one from the ‘global South’. ¹⁶ The impression is that Collings’ choice of these three case studies was very much informed by what Ran Hirschl calls the ‘prototypical cases principle’, where the cases under discussion ‘serve as exemplars of other cases with similar characteristics’. ¹⁷

Collings is obviously well aware also of the differences among the three experiences. The 1949 Basic Law of Germany and the Constitutions of South Africa (both the 1993 provisional Constitution and the 1996 permanent Constitution) are ‘never-again’ constitutions, as they define themselves ‘in opposition to an antecedent evil’ ¹⁸ (Nazism in Germany, apartheid in South Africa), and represent a response – in Kim Lane Scheppele’s words – to the previous ‘regimes of horror’. ¹⁹ Both the German Bundesverfassungsgericht and the South African Constitutional Court – as well as many other constitutional courts operating in post-authoritarian regimes (such as the constitutional courts of Central and Eastern Europe after the fall of the Communist regime) ²⁰ – presented themselves from the outset as the guardians of constitutions that celebrated and solemnised the end of illiberal rule.

The 1789 US Constitution, on the other hand, coexisted with slavery for several decades, and the Supreme Court tolerated it. ²¹ As underlined by Collings, ‘[t]he U.S. Constitution had been in force for 77 years when the Thirteenth Amendment abolished slavery and for 165 years when the Supreme Court [in Brown v Board of Education ²² ] condemned racial apartheid in public schools’. ²³ As a consequence, ‘[w]hereas the German and South African Courts could stake their early legitimacy on aggressive responses to the recent past, that posture was unavailable to their American counterpart. In the United States, the evil past was, in calamitous part, the Court’s own’. ²⁴ In Germany and South Africa the constitutional courts were ‘immaculate’ bodies that represented – in the eyes of the framers – a necessary instrument to mark a clean break with the previous illiberal regime. In the US, by contrast, the Supreme Court had long been ‘complicit in the evil’. ²⁵

Modes of judicial memory

Collings argues that, while referring to past evils, courts employ two modes of constitutional memory: the parenthetical mode and the redemptive mode. The parenthetical mode – which builds on the works by Edmund Burke and Benedetto Croce – considers the evil era as an exception to an otherwise worthy and unifying past. This period is viewed as a ‘baleful aberration’ ²⁶ from an otherwise noble constitutional tradition. Therefore, the constitutions adopted after an illiberal era do not represent a revolution. Rather, they are aimed to restore an older, glorious tradition. By contrast, the redemptive mode – which relies on the ideas of Thomas Paine – directly confronts the past evil and responds to it aggressively. The purpose of this mode of constitutional memory is not only to mark a clean break with the previous illiberal rule, but also to establish a new regime based on diametrically opposed principles and values. ²⁷

In an extremely careful and sophisticated analysis of the case law of the American, German and South African constitutional courts, Collings shows that all three bodies have used both modes of constitutional memory, but the US Supreme Court has mainly relied on the parenthetical framework, the South African Constitutional Court has employed particularly the redemptive mode, and the case law of the German Federal Constitutional Court has been characterised by a parenthetical-redemptive hybrid. A fascinating aspect of the book is that landmark judgments handed down by these courts – such as Brown, ²⁸ Luth ²⁹ and Makwanyane ³⁰ (but the list is obviously much longer) – are revisited under a different lens than the traditional one. Now the focus is not so much on how and why these decisions were crucial in guaranteeing the protection of people’s rights, but rather on how courts, through their rulings, looked at past evils, how they became agents of collective memory, and how they thus contributed to shaping the constitutional identity of the countries in which they operated. This perspective allows the reader not only to discover a new significance of these key rulings, but also to look at the history of the country in which the decisions are made under a different light.

Collings rightly emphasises the exceptionality of the parenthetical mode in the American context, which largely departs from its ideal type. Indeed, as also discussed above, US constitutional tradition did not pre-date slavery. Rather, there was for a long time a concomitance between the two. In this country, the parenthetical mode, which relies on the idea of continuity, ‘has proved particularly – irresistibly – attractive’. ³¹ The Supreme Court rejected, for example, to treat the Reconstruction Amendments (i.e. the Thirteenth, Fourteenth, and Fifteenth Amendments, adopted between 1865 and 1870) as the basis of a new, redemptive postbellum Constitution. Instead, the Court posited ‘a single, unitary constitutional tradition – interrupted, to be sure, by the parenthetical of slavery and secession, but resumed after a brief period of Reconstruction and a resounding closed parenthesis’. ³² The Supreme Court’s failure to sufficiently rely also on the redemptive mode, Collings contends, has prevented the country from fully coming to terms with its past. The consequence is that the ‘constitutional revolution [i.e. the ‘anti-slavery imperative’] so long remained – and, to an intolerable extent, still remains – unfinished’. ³³

In Germany, one of the most important decisions made by the Federal Constitutional Court during its first years of activity was Civil Servants. ³⁴ Despite the fact that the so-called ‘131 Law’ provided for a large reinstatement of Nazi-era civil servants who lost their posts when the War ended, and awarded generous back-pay, several of these civil servants were unsatisfied. In their opinion, they had never rightfully lost their jobs and wanted a full back-pay. The Court answered – in Colling’s words – ‘with a bombshell’. ³⁵ Indeed, the constitutional justices argued that the Nazi-era civil servants had no claims at all on the new Republic, and what they had received in terms of reinstatement and compensation only depended on the ‘benevolence’ of Parliament. It was not something that they were entitled to receive from a legal or constitutional point of view. In order to reach this conclusion, the Court largely relied on an historical analysis of the Third Reich, stressing that the civil administration had been completely penetrated and corrupted by the Nazi regime. The Court also made it clear that the new constitutional order would have been diametrically opposed to the previous one. This case is particularly relevant also because it sheds light on one of the most complicated issues in countries experiencing a transition from an autocratic rule, namely that of the continuity (or non-continuity) of the state, especially in terms of public officials. ³⁶

Among the numerous rulings of the German Constitutional Court analysed by Collings, it is also worth recalling those related to the constitutionality of political parties. The Basic Law grants the Bundesverfassungsgericht the power to declare the unconstitutionality of parties ‘that, by reason of their aims or the behaviour of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany …’ (Article 21(2)). It is above all by virtue of this provision (aimed to avoid the repetition of what was considered to be one of the decisive errors committed by the Weimar Republic, that is to tolerate ‘antisystem’ parties) that Germany became a ‘militant democracy’, to use the well-known expression coined by Karl Loewenstein in 1937. ³⁷ This constitutional provision did not remain a dead letter. In the first years of its activity the Court declared the unconstitutionality of a political party in two cases: in 1952 it ruled against a party of Nazi inspiration (the Socialist Reich Party), ³⁸ and in 1956 it dissolved the Communist Party of Germany. ³⁹ Collings rightly compares these two rulings with a judgment delivered in 2017, in which the Court decided not to ban a party of Nazi inspiration, despite the fact that its ideology was clearly incompatible with the Basic Law. The context made the difference. Indeed, as underscored by the author, ‘2017 was not 1956’. ⁴⁰ Now Germany is a mature, consolidated democracy that can tolerate antisystem parties that do not represent a concrete threat to the free basic order, whereas during the 1950s the country had just emerged from a totalitarian regime and therefore the need to preserve the democratic character of the new system was all the greater.

In its largely redemptive jurisprudence, the South African Constitutional Court played a key role in marking a clean break with the segregationist past. During its first phase, the Court mainly ruled on criminal cases, cases involving various forms of discriminations, and cases involving ‘negative’ freedoms (such as free speech), ⁴¹ but it subsequently also invoked apartheid memory in the contexts of socio-economic rights. ⁴² Interestingly, some of Court’s decisions have become ‘myths’, both within the country and abroad. This is the case, for example, with Makwanyane, ⁴³ in which the Court declared the unconstitutionality of the death penalty. Collings describes it as ‘an epic judgment’, in which the justices ‘didn’t just abolish the death penalty. They chanted a redemptive credo for post-apartheid South Africa’. ⁴⁴ Although the judgment was unanimous, all 11 justices wrote separate opinions, a fact that further strengthened the redemptive tone of this landmark decision.

It should be recalled that the role played by the South African Constitutional Court in the transition from apartheid to democracy was a unicum from a comparative perspective. Indeed, this body was even required to verify the constitutionality of the permanent Constitution. More precisely, it had to adjudicate on whether the final draft of the Constitution was in line with the 34 fundamental constitutional principles (contained in the 1993 interim Constitution) that were intended to guide the Constituent Assembly. The list of the 34 principles with which the final Constitution had to comply included, inter alia, constitutional supremacy, judicial review, an independent judiciary, the protection of the right to equality, separation of powers and checks and balances, protection of human rights, and a division of powers between national and provincial government. ⁴⁵ In the First Certification Judgment – which represents another example of a decision relying on the redemptive mode – the Court refused to certify the first draft of the Constitution and obliged the Constituent Assembly to implement certain changes in order to ensure compliance of the final version with the 34 principles. ⁴⁶ It was only with the Second Certification Judgment, which was handed down three months later, that the Court approved the new constitutional text. ⁴⁷ Interestingly, this role played by the Court gave rise to the ‘paradox of a constituted body co-participating in the constituent process’, ⁴⁸ marking the end of the idea according to which the pouvoir constituant is the only free and unlimited power. ⁴⁹

Which memory is better?

In the introduction Collings openly warns the readers that his book is mainly descriptive in nature (‘my aim is… to provide a thick description of how judicial memory has operated in some of the most influential and paradigmatic constitutional courts in the world…’ ⁵⁰ ), but while reading the volume one cannot help but also notice the normative dimension of the text. This ‘soul’ of the book clearly emerges in the conclusions, where the ‘normative instincts’ ⁵¹ of the author come to the forefront. Overall, the descriptive and normative dimensions are well balanced. Collings first describes the different forms of judicial memory and then shows how they operated in practice. For each case study, he identifies strengths and weaknesses of the constitutional memory employed by the respective Court. This allows the reader to form his or her own idea about these different approaches and thus be able to better evaluate the author’s point of view. The impression is that Collings is not so much interested in convincing the reader to share his opinion on which judicial memory is better, but rather in prompting the reader to reflect on the consequences of choosing one memory over another.

Collings calls for a sort of merge between the parenthetical mode and the redemptive mode. In his opinion, the fusion between these two frameworks is the ‘ideal approach’. ⁵² Indeed, according to the author, ‘it is dangerous to look at a tradition without its evils; it is also dangerous to look at evils without a tradition’. ⁵³ Collings also argues that even if the two modes should be mingled, the redemptive approach needs to come first: ‘There is a need to remember and redress, followed by a need to remember and move on.’ ⁵⁴ As recalled above, out of the three courts considered in the book, only the German Bundesverfassungsgericht consistently (and successfully, in Collings’ view) relied on a combination of the two modes of constitutional memory. Indeed, the Court first responded aggressively to the Nazi regime, but over time it also drew abundantly on the pre-1933 constitutional tradition. Collings suggests that the combination of the two modes is more likely to ground the constitutional identity of a country and to find broad acceptance in the population, especially in the long run. ⁵⁵

By contrast, in South Africa, the largely redemptive jurisprudence of the Constitutional Court contributed, inter alia, to strengthening the perception that the Constitution was the solution to many daily problems of the citizens, a fact that often generated a sense of frustration among the population every time the promises contained in the constitutional text were not maintained. Furthermore, the redemptive jurisprudence, which is mostly negative in nature (being in opposition to past evils), might hinder the construction of the state’s positive identity. As Collings puts it, ‘redemptive memory is more powerful as a destructive than as a constructive mode’. ⁵⁶ The author also warns that, if untampered, redemptive jurisprudence might be (ab)used by courts to justify a permanent judicial activism. ⁵⁷ The parenthetical mode is equally (or maybe even more) problematic. Collings is right when he argues that in the American context, the dominance of this mode of constitutional memory in the Supreme Court’s case law ‘long prevented and still hinders a full reckoning with the [country’s] towering evils’. ⁵⁸ Indeed, ‘noble traditions cannot be resumed until towering crimes have been squarely confronted’. ⁵⁹

The relevance of the parenthetical mode and the redemptive mode obviously transcends the three case studies on which the book focuses. Thus, for example, in an interview given to the Italian newspaper Il Corriere della Sera in which he discussed his volume, Collings masterfully applied these modes of constitutional memory to the case of Italy:

Parenthetical memory is important because it is difficult to construct a political identity without some positive historical foundation. For example, one could call post-World War II Italy an ‘anti-fascist’ Republic, but this is a purely negative foundation. You also need something else – Dante Alighieri and Francesco Petrarca, for instance, the masters of the Renaissance and the liberal ideas of the Risorgimento. Being Italian is a problematic identity if it only means not being fascist. On the other hand, one cannot simply set aside the fascist era as an exceptional divergence – ‘a parenthesis’ – in the course of the history of Italian liberalism.⁶⁰

Collings, who gave this interview in January 2021, anticipated a heated debate that would erupt about a year and a half later, following the September 2022 elections, with the establishment of the most right-wing government in Italy’s post-World War II history. Indeed, the clear victory of Giorgia Meloni, leader of Fratelli d’Italia (a party that has its roots in the Movimento Sociale Italiano, a neo-fascist-inspired party), ⁶¹ immediately channelled the political debate into questions such as: What does it mean to be a fascist? Is the current government to be considered neo-fascist? Or post-fascist? Or neither? Can fascism and communism be equated? Is it necessary for a politician to publicly declare him/herself anti-fascist? These questions – which were often the subject of instrumentalisation by all political forces – are extremely controversial in todays’ Italy, ⁶² a fact that shows that the country has yet to fully come to terms with its past. But if reckoning with past evils is delayed, Collings warns in the same interview, ‘the debts increase with interest’. ⁶³

What the book does not say

There is only one aspect of Collings’ book that has left me somewhat puzzled, an aspect related to what the book does not say. In this volume (and also in a subsequent work), ⁶⁴ Collings rightly stresses that constitutional courts often invoke historical evil as a powerful legitimating resource. This was certainly the case of constitutional justices in the US, Germany and South Africa, whose appeals to constitutional memory were a very important way to strengthen their decisions. What the book could and should have better explained is that many other courts in the world decided not to rely on history in their decisions, even in those cases which were strictly connected with the legacies of the previous illiberal regime. In this respect, the case of Italy is emblematic. In this country, most of the key judgments dealing with the country’s authoritarian past were based on purely legal reasoning, technical considerations with little scope for historical or political analysis. The example par excellence is the judgment of the Italian Court of Cassation of 7 February 1948, relating to transitional justice. In that case, in upholding prosecution of Fascist crimes notwithstanding the ban on retroactivity (provided for by Article 25 of the 1948 Constitution), the Court introduced for the first time a distinction between ‘programmatic constitutional provisions’ (i.e. provisions that exist only as ‘programs’ for the legislator, and which thus could not be immediately enforced) and ‘preceptive constitutional provisions’ (i.e. provisions that could be immediately enforced), and deemed Article 25 to be programmatic. Historical references to the previous Fascist regime were entirely lacking. ⁶⁵ Not even in the Constitutional Court’s first decision (no. 1 of 1956, a truly landmark judgment), in which for the first time the Court struck down a piece of legislation enacted under the previous dictatorship, had the judges made historical considerations relating to the Fascist regime. ⁶⁶

Among the various reasons that explain the different approaches towards the use of history in constitutional interpretation, one might include the composition of the courts. In Germany and South Africa (but also in other countries), ⁶⁷ constant references to past evils can also be explained by the fact that the first constitutional courts’ judges personally symbolised the repudiation of the previous illiberal regimes. ⁶⁸ Albie Sachs, who was a justice of the Constitutional Court of South Africa from 1994 to 2009, is probably the best-known example. ⁶⁹ In Italy, by contrast, the continuity with the past was much more evident, also because the purges of public officials (including judges) produced only limited effects. Not even all members of the first Constitutional Court were above suspicion. Indeed, the authoring judge who drafted the first judgment of the Court was Gaetano Azzariti, who played a leading role in the Fascist administration: he was the head of the Legislative Office at the Ministry of Justice for more than 20 years, taking part in the drafting of Fascist legislation, including the racial laws, and was appointed president of the Racial Tribunal (Tribunale della Razza) in 1939. In spite of his involvement in the regime, in 1955 he was appointed by the President of the Republic as a Constitutional Court judge, and it was actually Azzariti (and here the paradox is astonishing) who drafted Judgment no. 1 of 1956, declaring for the first time the unconstitutionality of a Fascist-era piece of legislation. ⁷⁰

I obviously did not expect Collings to add as a separate case study a country whose constitutional court had opted for a very different approach compared to that followed by the US, German, and South African counterparts. This would have likely shifted the focus of the book. ⁷¹ However, regardless of what the advantages and disadvantages of this judicial approach may be, some references (maybe in the introduction or in the conclusions) to the constitutional courts that have decided not to put historical evil at the heart of their case law would have offered a broader and more complete picture of courts’ action in transitional settings, and would have contributed to better clarifying that invocation of past evils is not necessarily a standard practice.

The second book of a trilogy?

Collings’ book is solid from a methodological point of view, thorough in its analysis, convincing in its arguments, and original in its approach. These qualities can hardly be found in works dealing with such an extremely complicated issue as the relationship between law and time. It is true that work on memory never ends and that questions related to the past have no obvious answers. Scales of Memory, however, undoubtedly has the great merit of facilitating this work and helping to provide answers. Collings shows that there are no easy solutions, but clearly indicates what are the consequences of choosing one type of memory over another.

In order to fully appreciate this book, it should also be emphasised that Scales of Memory is not an ‘isolated monad’, but fits within a well-defined academic project. In an article published in 2017, ⁷² Collings proposed three possible approaches to the study of comparative constitutional history. The first approach, called ‘perspectival history’, focuses on a single jurisdiction. It will be comparative ‘thanks to the author’s unique perspective and intended audience’. ⁷³ The prototype of this first approach is Tocqueville’s Democracy in America: thanks to his external perspective, Tocqueville produced a (mainly implicit) comparison between France and the US. Collings’ first book, Democracy’s Guardians. A History of the German Federal Constitutional Court 1951–2001, ⁷⁴ is another example of perspectival history. Indeed, in examining the history of the Bundesverfassungsgericht, Collings was writing not only ‘as an American’, but also ‘as a comparative constitutional historian’. ⁷⁵

The second approach to comparative constitutional history, which is called by Collings ‘thematic history’, focuses on a single theme (that might be of a doctrinal, institutional, procedural, or theoretical nature) across different jurisdictions, following a diachronic analysis. ⁷⁶ This is where Scales of Memory is situated. Indeed, as discussed in this review, Collings’ second book compares three case studies ‘that are contrasting, distant, and diachronic’. ⁷⁷

The third approach to the comparative study of constitutional history, which is defined by Collings as ‘relational history’, examines ‘multiple jurisdictions within a single chronological frame’, attempting ‘not merely to compare the jurisdictions with one another, but also to identify mutual influences, both reciprocal and external’. ⁷⁸ Collings does not provide for a hierarchy among the three approaches to comparative constitutional history, but he sees relational history as the ultimate goal to strive for: ‘I believe that some of the best work in our field will be relational; but relational histories … will have to build on the foundation of previous perspectival and thematic work’. ⁷⁹ Specifically, Collings’ intention would be that of writing ‘a portion of the history of global constitutionalism on an integrated, relational basis’. ⁸⁰ This would represent the third book of a ‘trilogy’ on comparative constitutional history. Expectations can only be sky high.