A measure of last resort: Pseudo-constitutionalism and the persistence of a self-restraint slogan in Argentina

Abstract

This article addresses the bastardization of constitutional law in Argentina and the corrosive power of legacies of authoritarianism. It offers a genealogy of the use by Argentina’s Supreme Court of self-restraint canons from the time when they were borrowed from the US Supreme Court in the late nineteenth century. Partly resulting from the country’s experiences with military rule, the court transformed or expanded these canons, which entailed a gradual depreciation of statutes as the (uneasy) cornerstone of constitutionalism. Based on a fresh dataset and employing narrative and network analysis, the article focuses on a slogan the court has invoked since the 1960s: invalidating a rule is a matter of extreme institutional gravity and hence a strategy of last resort. Under the 1976–83 dictatorship, the court applied the slogan to various rules, including those passed by the military. It thus invoked familiar canons outside its scope conditions, conveying an illusion of constitutional regularity by masking the abnormal in acceptable garb and contributing to the regime’s legitimation. While the democratic court abandoned the most blatant expressions of authoritarianism, connections persisted, manifesting in the frequent citations to the dictatorship court’s use of the slogan and its extension to any rule. Authoritarian legacies die hard.

I. Introduction

For their weapons are the same as those we have become acquainted with in the absolutism that is disguised with a pseudo-constitutionalism, namely, the stating of things that are not so.

– Ferdinand Lassalle (1927: 47)

Suppose an uninformed person took a glance at the spines of the bound volumes in which the Supreme Court of Argentina, the body situated atop the country’s judiciary, publishes its judicial and administrative decisions. The reports have been appearing with delays but without interruption since the court’s first decision in October 1863. Most likely, the person would not have any inkling about the much-convoluted performance of the country’s political system during the twentieth century. Six coups led by the then all-too-powerful armed forces put an abrupt end to the administration of elected presidents and closed down Congress’s doors, and intra-elite splits led to changes within the ensuing de facto administrations. State-sponsored violence became all-pervasive. And presidents with electoral credentials ruled amid dramatic political exclusion and/or rule of law shortcomings.

The onlooker would not know either of the forced turnover of all or most of the Court personnel after three of those coups and subsequent episodes of restoration of electoral rule. The Court was one of the targets of the coups – a spoil of war – as much as a source of legitimation of them. It enjoyed a degree of autonomy until the first decades of the twentieth century, but in 1930 it validated the first coup in an ex-officio decision outside the scope of a case, grossly exceeding its jurisdiction. In the late 1940s, during the Perón administration, four of the five justices were subject to impeachment, which resulted in one resignation and three removals. The new justices would in turn be ousted after the 1955 coup when another set of justices was selected. Only some of these were kept a few years later, and in 1966 a new coup put this cycle in motion again. Without entirely disappearing, this cycle would diminish with the return to electoral rule in 1983.

These changes greatly affected the court’s independence and influenced its case law in significant respects. Despite this severe personnel instability and a series of irregular appointments, though, the court remained in operation throughout. The court – and the judiciary more generally – were the only branch that gave a whiff of appearance of performing regularly. The court went on with its daily business almost without breaks and it remained practically unchanged in terms of its main operational traits: its jurisdiction, the rules on how to reach it, and its modes of deciding. Unlike the Praetorian powers of the executive under dictatorship and the vacuum left by the closure of Congress, the seeming reason-giving nature of the court’s function may have helped radiate a minimum image of continuity even against the backdrop of utter instability.

This modicum of regularity emanated by the court’s institutional presence may have served some rule-of-law values, such as guaranteeing the administration of justice to litigants, which is not to say that ordinary justice in the country has been satisfactory or timely. At the same time, however conscious the justices were, it may have had a deleterious effect in legitimizing what was abnormal and normalizing a travesty of constitutionalism (Barrera 2015: 331; Bohoslavsky 2015: 27–30; García-Mansilla 2013: 98). This article studies one particular example or manifestation of this phenomenon. It focuses on a familiar set of doctrines embraced and asserted by the court across the periods centred around its self-restraint vis-à-vis the other branches and connected to the presumption of constitutionality of legal rules. According to the court, invalidating a rule is an act of ‘extreme institutional gravity and must be considered the last resort of the legal system’. Since serious consequences attach to an invalidation, courts must refrain from this if it can be avoided by any means. A rule’s unconstitutionality must be established without any doubt before it is struck down.

These notions resonate well with a set of canons of self-restraint developed in the nineteenth century by the US Supreme Court and defended by the likes of James B Thayer in relation to statutes. The similarity is not coincidental since, at the dawn of its operation, the Argentine Court liberally borrowed from, or relied on, its North American counterpart – its model (Miller 1997). The canons’ seeming driving force is an array of concerns that are often not fully articulated and that give shape to the presumption of constitutionality. First, to paraphrase Waldron (1999), dignity must be ascribed to legislation as an expression of the people’s will, so the latter must be preserved whenever it is normatively possible. Not all rules are created equal (or equally), and rightly so. Second, democratic legislators should not be deemed oblivious to constitutional constraints (Thayer 1893: 155–56; Tushnet 1999: 57–58). Third, courts must be aware of their democratic, institutional and epistemic shortcomings.

These deferential doctrines are recognizable in the United States mainly in the sphere of statutory interpretation, although in certain respects they have been stretched to administrative regulations and action. Instead, the Argentine Court has invoked them liberally and uncritically regarding all types of rules, including arcane administrative action and, in many cases, rules introduced by de facto military administrations. Although regrettable, this was perhaps expected of a court serving under a dictatorship. But the partially confounding use of the doctrine also took place under electoral rule and, to some extent, has been preserved right up to the present day.

This article traces the uses and misuses by the court of these self-restraint canons and concomitant language since they were first invoked in the nineteenth century and the specific employment of the ‘extreme institutional gravity, last resort’ slogan since it was first cited in the early 1960s. To better assess the continuity of uses across the periods, the article also employs network analysis. The mismanagement of imported doctrines partly speaks to the inherent dangers of borrowing from foreign jurisdictions (Dixon and Landau 2019). Good or bad, the doctrines were articulated in the American context within a particular scope and with some (though sketchy) justification. The Argentine Court apparently implemented the doctrines following that scope and justification, but at a later stage it expanded and distorted them without much, or any, explanation – likely benefitting from the legitimacy of the US model.

The confusing use of these doctrines likely reflected the comparatively low value that statutes were assigned in the legal and political system after a seemingly never-ending cycle of institutional breakdown, political exclusion and violence. The judicial endorsement of the myriad new regulations passed by dictatorships, as well as the reinforcement of executive powers in the country, for pragmatic, ideological or strategic reasons, both expressed and magnified the loss of saliency of statutes and their confusion with any other legal rules, no matter how faulty their pedigree. By asserting a similar presumption of constitutionality of all rules without a convincing rationale, the court at times betrayed its role as constitutional guardian and engaged in unacceptable subterfuge (Calabresi 1982), approaching the disguise that, according to Lassalle, was taking place in Prussia in the hands of the executive.

The court preserved the ‘extreme institutional gravity, last resort’ slogan after the restoration of electoral rule in 1983. On a few occasions, it underscored the singularity of statutes, in one of several signs of progress with respect to the past. But it also extended the slogan to other rules, including rules issued by the court itself. And it may have done so unreflectively, citing an old and tired adage to reinforce the reasoning behind a decision. This is perhaps a sign of availability bias, but the indiscriminate availability of the adage speaks to the bastardization of constitutional law in the country and the corrosive power that legacies of authoritarianism may have.

By unduly extending a veneer of legitimacy, the court’s justification in some of the cases it chose not to invalidate was faulty and may have had the result, if not the aim, of legitimizing rule-making powers where this was not merited. However, the article should by no means be interpreted as a plea for heightened judicial assertiveness. It is entirely possible that some or most of the rules under consideration did not present serious constitutional objections and were correctly upheld. Moreover, the article is far from asserting that the judiciary, under either dictatorship or democracy, inserted the slogan to duck every challenge of unconstitutionality. Even under dictatorship, the court occasionally struck down legal rules in their application to cases and it did not always concern itself with self-restraint. Indeed, the claim that statutes were devalued is compatible with a potential finding or interpretation that, every now and then, the court wrongly invalidated them – an issue that is outside the scope of this article. In any case, words matter, and a court should justify its decisions properly and without the use of artifice.

By exploring judicial decision-making under authoritarianism, legal continuities and discontinuities across regime types, and the normalizing power of courts, this article aims to contribute to the literature on courts in authoritarian regimes and new democracies. The article is structured as follows. Part II examines the legitimizing role of courts under authoritarianism. Parts III and IV discuss self-restraint as developed by the US court and the Argentine court, respectively. Part V analyses the uses and misuses of the slogan under dictatorship and democracy, and a brief section concludes the article.

II. Normalizing pseudo-constitutionalism

In a famous speech from 1862, Ferdinand Lassalle (1927) denounced the ‘pseudo-constitutionalism’ of the Prussian government. Under the leadership of von Bismarck, King Wilhelm I had disguised the exercise of naked power behind the façade of constitutional language. For Lassalle (1927: 46), it was time to force absolutism ‘to take off its mask’.

Stating ‘things that are not so’ is a usual expedient of authoritarian regimes, either as a defensive shield – as in the statements of its officials – or a potential source of legitimation – as in the pronouncements of courts. While the former can be cast aside as empty parlance, the judicial subterfuge of courts under authoritarianism can be more devastating in the concealment, and at times validation, of what is a collapse of democratic and constitutional governance. It can also extend its deleterious effects to succeeding periods of democratic rule. Argentina provides an example of this phenomenon.

Legitimation and instability

Despite their shortcomings, ‘abusive’ constitutional regimes (Landau 2013) usually possess a central feature that dictatorships normally lack – a degree of accountability and social legitimacy. It is no wonder that, as Moustafa and others have remarked, both courts in dictatorships and the use of a ‘rule-of-law rhetoric’ become convenient vehicles to make up for part of that missing legitimacy (Moustafa 2014: 287; Snyder 1984). Now, to spill legitimacy over an authoritarian regime, courts need to have it in the first place. This can hardly happen if they are seen as an exercise of naked power. This is one reason why a dictatorship may decide to leave courts untouched, as transpired in the Philippines after Ferdinand Marcos’s power-grab in 1972 (Del Carmen 1973; Moustafa 2014).

Dictatorships typically rule with (although not necessarily under) a constitution of some kind. Unless it is an utterly repressive constitution, it may provide some space for contestation in courts – perhaps as a leftover from the previous regime. Moustafa (2014: 287) rightly says that courts under authoritarianism have an ambiguous role ‘as instruments of political control’, but also by ‘opening new (albeit limited) avenues … to challenge the state’. ¹ But because they fear for their job, owe their post to the new regime or share its aims, judges will likely recognize the authorities and mostly rule in their favor, as has happened around the globe (e.g. Mahmud 1994; Moustafa 2014). Apart from the open recognition of the regime, however, courts may convey an illusion of constitutional regularity by masking the abnormal in acceptable garb, creating some space for ‘pseudo-constitutionalism’ and legitimation. A court’s tweaking of old doctrines and conceptual categories to address the new reality may help transmit such an image of normalcy.

This is the trajectory the Argentine Supreme Court largely followed, thus becoming a strong candidate to provide a modicum of legitimacy to authoritarian regimes even if its justices were repeatedly sacked and replaced (Snyder 1984: 519; Groisman 1987: 8; Barrera 2015: 328–331; Bohoslavsky 2015: 30). Institutional breakdown and instability led to the court’s open ratification in a handful of cases of the new irregular situations. In countless other cases, though, the court outwardly fitted that reality into old moulds, thus normalizing it. The head of the country’s judiciary in operation since 1863, the court was targeted and its personnel were removed several times in the twentieth century, mostly through forced resignations following both coups and a subsequent return to electoral rule (Castagnola 2019).

Though it showed a pro-regime stance, the court enjoyed a degree of autonomy until the first decades of the twentieth century, as many authors have described (e.g. Bill-Chavez 2004; Castagnola 2019; Helmke 2004). In 1930, it granted its imprimatur to the first coup (against President Yrigoyen) in an ex-officio decision. It would similarly endorse the coups of 1943, 1955, 1962 and 1976. In 1947, months into Peron’s first administration, four of its then-five justices were subject to impeachment, which resulted in one resignation and three removals and the selection of partisans. The newly appointed justices would in turn be ousted in 1955 when another set of justices were selected. Only some of these were kept a few years later; in 1966, a new coup put this cycle in motion again – the removal of all justices and a set of appointments, another turnover in 1973, another after the 1976 coup and yet another with the return to democracy in 1983. In some cases, the instability radiated to the lower courts.

This merry-go-round would diminish without disappearing with the stabilization of electoral rule – or the advent of democracy without restrictions – in 1983. In 1990, the incoming President Menem was able to pack the court with his cronies after expanding its personnel. This was followed a decade later by some impeachments and resignations, and a new set of appointments – of comparatively more respected candidates – by the Kirchner administration. These appointments and many of the court’s ensuing decisions, including those that allowed for the reopening of criminal investigations for human rights crimes during the dictatorship, did much to rebuild social support for it. By the end of the 2010s, however, the court could not escape a polarized political environment. At the time of writing, another bill was being debated to expand the number of justices sitting in the court – currently composed of five members. ²

Considering those justices appointed after the 1947 impeachments up to 1983, the average tenure of a nominally life-tenured justice was four years. It was ten years before 1947 and eleven years since 1983 without considering those serving as of 2020. ³ Given this, the available literature rightly points to the court’s radical instability during the twentieth century (e.g. Bill-Chavez 2004; Carrió 1996; Castagnola 2019; Helmke 2004; Oteiza 1994). This affected the court’s independence, compromised its role as a constitutional custodian and influenced its case law.

The extreme personnel instability notwithstanding, the court remained in operation under mostly unchanged rules, exhibiting a degree of continuity of the institution. It went about its daily business of dealing with cases almost without breaks and preserving its main operational traits – its jurisdiction, the rules on how to reach it, and its modes of deciding. Together with the seeming reason-giving nature of its function, this continuity may have helped to communicate an image of regularity across periods (Bohoslavsky 2015: 30; Osiel 1995: 524; Barrera 2015: 328–331). A sense of stability may also have resulted from the fact that the court – like its model, the US Court – is embedded in the ordinary judiciary. Unlike a constitutional court, which either sits outside the ordinary judiciary or operates following a logic different from that of the ordinary judiciary, which means it can therefore be more easily limited (Ferreres Comella 2009), the court is an indispensable part of that judiciary. It escaped no one that, starting in 1955 and during dictatorships, the new members of the court were appointed by a de facto administration by sidestepping the Constitution. That court was a dictatorship court, populated by technical experts who often were vehement anti-Peronists. But compared with the visibly Praetorian powers of the military executive and the absence of a Congress, the court was the only major federal institution conveying a minimal appearance of performing in a manner close to that prescribed in the Constitution.

Regardless of whether the court’s justices intended to relay this image of regularity – given that intentions should not be read from outcomes, and even less so in the case of a collective body – the image was a good candidate to serve as a source of legitimacy for the military authorities. The court’s endorsement of military coups clearly entailed the validation, if not the exercise, of naked power (Diana 2012, 2013: 46; García-Mansilla and Gascón 2012; Prat 2015). Yet, despite their extraordinary importance, those decisions coexisted with tens of thousands of routine decisions in which the court displayed an array of tests and criteria that were recognizably familiar to the legal community, in many cases to uphold the regime’s actions (Groisman 1987: 8). ⁴ In the view of some scholars, the legitimizing nature of judicial decision-making may even have extended to decisions that were seemingly against the regime, a matter beyond the scope of this article. Scholars including Snyder (1984), Osiel (1995) and Bohoslavsky (2014) have investigated the judicial response to state terror during the country’s last dictatorship (1976–83) and the extent to which a handful of ostensible acts of confrontation with the regime involved merely rhetoric to create the idea that ‘the rule of law was being upheld’ (Bohoslavsky 2014: 30) or had actual limitations. ⁵

It follows from the preceding paragraphs that this sense of normalcy emanating from the court can be a double-edged sword, not only satisfying legal certainty and predictability as is often stated about comparable scenarios. Institutional breakdown led to the judicial support of multiple irregular legal figures, and the sense or illusion of continuity may have helped to mask some of these irregularities, defended under the guise of traditional doctrines and canons. This article deals with one such set of doctrines regarding self-restraint. The doctrines were frequently misapplied under dictatorship but continued to be used under democracy. In the period inaugurated in 1983, the court eliminated some of their most problematic aspects, but not all of them, and not entirely. This would have required a kind of conscious distancing from the previous period that has so far only partially occurred.

This account of authoritarian judicial legacies complicates our understanding of institutional stability. In periods of transition after prolonged breakdown, a renewed insistence on an old institution and practices, while undoubtedly possessing virtues, might also hinder the type of fresh start or open interrogation that a society may need, free from the inertia of the past. Leaders of some democratic revolutions, including post-World War II democracies, did not hesitate to overcome the existing judiciaries and create new adjudicatory bodies to prevent those judiciaries from extending their gravitational pull over the new regimes. Although in those cases the fresh start was motivated by the fact that the pre-transition judges would still have sat under democracy, perhaps something analogous can be said about the courts themselves – their inertia, traditions, procedures and canons.

In Argentina, it was admirable to express a new commitment to the court when democracy was restored in 1983 since it was an important symbol of constitutional regularity, not just authoritarianism. With ebbs and flows, the democratic court appointed that year left most of the authoritarian legacy behind, and this trend received a boost first when an amendment in 1994 breathed life into its decision-making by placing human rights treaties on par with the Constitution and later when new justices joined it in 2003–05. The court today is light years away from embracing that legacy and is a vital (though understandably often disparaged) part of the nation’s democracy. Yet some traces of that legacy could be detected in the 1980s and 1990s and, albeit much more subtly, can still be detected today.

Open confirmation and subterfuge

Apart from its open endorsement of dictatorships, which included the period between 1976 and 1983 during which the junta in power systematically tortured, executed and/or ‘disappeared’ tens of thousands and appropriated newborns from their captive mothers, the court on occasion recognized their ‘revolutionary’ constituent powers (e.g. Ortiz 1958; Molinas 1968). The dictatorships nominally kept the 1853 Constitution in place, but only to the extent that it did not contradict their set objectives. The court increasingly upheld the regulatory power of military administrations. Under the label of ‘de facto doctrine’, scholars for decades would discuss the validity of those rules under both dictatorship and restored democracy (e.g. Diana 2013; García-Mansilla and Gascón 2012; Nino 1985).

In 1930, maintaining that the new administration had promised to respect the Constitution, the court both recognized and placed limitations upon those powers, which could not ‘ignore’ constitutional protections (Acordada 1930). Three years later, it validated them again while further constraining their scope (Malmonge 1933). The de facto executive, it said, enjoyed the powers of a regular executive so it was barred from legislating. In 1947, however, the Perón Court considered the rule-making powers of the de facto executive a natural consequence of the exercise of power, equating military orders with statutes and lifting whatever condition it may have imposed previously (Arlandini 1947). With a brief interlude in 1973–76, this doctrine would subsist until 1983, when the new democratic court in its third published decision subjected the rules’ validity to Congress’s explicit or implicit recognition (Aramayo 1984). ⁶

Given the salience of this open confirmation, it is understandable that the literature has paid more attention to it than the mundane application of seemingly traditional tests and doctrines. Extensive criticism, for example, has been levelled against the court’s legitimation of coups by the employment of judicial discourse to give cover to the illegitimate and ‘justify what is legally unjustifiable’ (e.g. Cayuso and Gelli 1988; Diana 2012, 2013: 46; Prat 2015; Volosin 2018). However, that blanket recognition coexisted with somewhat subtler approaches denoting normalcy.

The court employed tested ideas and canons of constitutional law and in some cases distorted them to fit a new reality. For example, the fleeting post-1955 coup court unsurprisingly recognized the military authorities who had appointed it, and its justices distanced themselves symbolically from their immediately preceding colleagues. But they did so in part by invoking familiar constitutional notions, thus obfuscating the existing irregular situation. They cited the separation of powers to strike down rules and overturn decisions from the Perón administration (e.g. Mouviel 1957), something that may have been appropriate under ‘normal’ times but that the court hesitated to apply to the sitting authorities, which governed without a Congress.

The court’s credentials under dictatorship were as flawed as those of the administration. But the court concerned itself with occupying an ambiguous role between the normal and the abnormal and, as an institution, it straddled the regimes. Despite its irregular character, the court exuded regularity (Barrera 2015: 330). The doctrines explored in this article precisely involve the court’s internalization of familiar constitutional ideas, which are then adjusted to deal with a new reality and serve new purposes. After repeated use under a still recognizably familiar form, these doctrines became routinized and incorporated into the pantheon of judicial canons.

The role of statutes

Not all rules are created equally in a constitutional democracy, and statutes should be assigned the highest place under the constitution. Statutes have had an uneasy standing within constitutional theory because the power of majorities has had an uneasy standing within it (Waldron 2016). Yet few would deny their place in representative democracies as the (however imperfect) expression of people’s will. Waldron has famously defended the ‘dignity’ of legislation, although how much dignity can be ascribed to statutes will depend in part on the strength of democratic institutions. While, pace Waldron, statutes should not at all be immune from judicial scrutiny, a long-lasting polemic has reminded of the difficulties – including the ‘counter-majoritarian difficulty’ – involved in this enterprise and the caution that, according to many scholars, judges must express when dealing with them. These considerations are not so forcefully present in the case of other rules, not to mention rules passed by a military regime.

Due to regime breakdown in Argentina and its acquiescence by a portion of society, for decades statutes were not assigned special saliency. And, given political exclusion and rule of law and accountability deficits during elected administrations, judicial deference towards them may often not have been as clearly justified as it may be in more consolidated democracies. But this does not mean that rules that were not statutes must be treated as such. For all the shortcomings of the country’s democracy, statutes arguably still possessed greater legitimacy than other rules. Through a series of doctrines and canons, the court nonetheless seemed to extend to the latter the same level of deference it accorded the former. In some respects, it still does.

This amalgamation of statutes and other rules partly stems from the court’s unjustified endorsement of the military regimes’ law-making powers, but the canons explored in this article reach beyond this endorsement. It is one thing to say that certain utterances in the form of directives are initially deemed official rules even if made by a dictatorship, but it is a different thing to extend to them, as the court often seemed to do, a blanket presumption of constitutionality. Both manifestations are akin to Lassalle’s (1927: 47) pseudo-constitutionalism, or the stating of ‘things that are not so’, and when a court charged with guarding the constitution engages in such discourse, it risks undermining its important mission. ⁷ Among other scholars writing on the issue, Diana (2012, 2013) has similarly denounced the court’s ‘pseudo-scientific justification’, ‘pseudo-institutionality’ and ‘metajuridical’ discourse when legitimating the coups and the initial validity of rules, regardless of their origin. Still, when a court applies a tested doctrine such as the presumption of constitutionality outside its normal scope conditions, it seems to engage in a more subtle type of subterfuge by wrongly conveying a sense of regularity.

Individuals, businesses and officials must consider legal rules valid until struck down if they have the appearance of having been duly enacted. ⁸ And a court can naturally demand from a rule challenger that they support their arguments. In employing doctrines of self-restraint and presuming the constitutionality of a legal rule, however, a court often does something more, as will be discussed in Part III: it announces an active disposition to regard the rule as in line with the constitution. To the extent that fundamental rights are not visibly implicated, this a court can say meaningfully of a democratic legislature due to weighty normative considerations (Metzger and Morrison 2013, 1729) and, in certain cases, of some administrative action. But a court has hardly any reason to extend this deferential attitude or language to all authorities – including a de facto military government – under any circumstances, and less so without justification. ⁹ Nevertheless, the court has done so on occasion, under both dictatorship and democracy.

The remainder of this article offers a genealogy of the court’s reception, expansion and internalization of self-restraint canons. It also traces the origin, uses and misuses of a slogan it has regularly invoked to encapsulate and accompany those canons. The exploration is purposely biased in that it only considers the court’s assertions of self-restraint in dismissed challenges to legal rules. In parallel, the court has struck down many legal rules in their application to the cases under examination. The article’s claim, as noted in Section I, is not that the court has been excessively deferential throughout. The article focuses instead on the court’s employment of the canons as mechanisms that often reveal the use of unjustifiable subterfuge and show some misguided continuities across the periods.

The article’s exploration is also qualitative rather than quantitative. The court’s use of the slogan is not conspicuously present in the docket. Such assertions have been sporadic, an average of three to five a year under dictatorship or democracy, while the court issues thousands of decisions a year. Still, they illustrate well the persistent power of authoritarian legacies and the potentially legitimizing force of judicial discourse.

III. Introducing self-restraint

For reasons including constitutional restrictions and normative and prudential considerations regarding its role within a democratic system, the US Supreme Court developed over time a series of doctrines of self-restraint (e.g. Katyal and Schmidt 2015; Nelson 2015). While that court has likely wavered in the application of them and it has not been free from criticism, a discernible pattern can be articulated with concomitant rationales. It will be shown in Part IV that some of these doctrines were replicated (and then partially distorted) in Argentina, so it is useful to analyse them, as well as the general background against which they operated. Therefore, this is partly a story of borrowing gone wrong. It resonates with recent work discussing non-innocent, or strategic, uses of borrowing (Dixon and Landau 2019), albeit with some nuances. The main finding will not be that the Argentine Supreme Court misappropriated stable and spotless doctrines, since the doctrines have been both in flux and debated also in the United States. Rather, it will be that the Argentine Court stretched the doctrines beyond their most straightforward foundations while offering an even sparser justification than its North American counterpart, at times creating the sense that it used their outward logic as camouflage for legitimation.

Importantly, this was the result of a gradual evolution in an endemically unstable milieu. This evolution may have also taken place had the doctrines entirely originated in Argentina, which explains why this was not a case of a ‘malicious legal transplant’ (Siems 2018). But, at least in the first steps of this transformation, the court likely benefited from the legitimacy of the US model, which helped it to radiate a wider sense of normalcy in what it was doing. ¹⁰

Avoidance in the United States

In 1936, in his famed concurrence in Ashwander (1936), Justice Brandeis condensed seven ‘rules under which [the Supreme Court] has avoided passing upon a large part of all the constitutional questions pressed upon it for decision’. These had been interspersed into the court’s decision-making before (Kloppenberg 1994; Nolan 2014). ¹¹

Some concerned mootness and justiciability (Kloppenberg 1994; Nolan 2014). The court would not decide upon the constitutionality of a rule in a non-adversary proceeding and would not ‘anticipate a question of constitutional law in advance of the necessity of deciding it’ (Liverpool 1885). Moreover, any person challenging a legal rule must show that they are injured by its operation. According to what is usually known as the ‘last resort rule’, the court would not ‘pass upon a constitutional question’ if there is ‘some other ground upon which the case may be disposed of’ (Ashwander 1936). If the question could be dealt with on non-constitutional grounds, including state law, this must be preferred. The multiple cases embracing this notion include Siler (1909), Light (1911), Hagans (1974) and Westcott (1977) (Kloppenberg 1994).

In Rescue Army (1947), Justice Rutledge enumerated some motives underlying these and related rules:

They are found in the delicacy of that function, particularly in view of possible consequences for others stemming also from constitutional roots; the comparative finality of those consequences; the consideration due to the judgment of other repositories of constitutional power concerning the scope of their authority; the necessity, if government is to function constitutionally, for each to keep within its power, including the courts; [and] the inherent limitations of the judicial process …

The previous rules or canons usually pertain to any official action including statutes, but also executive orders and administrative action. But others only apply to statutes, chiefly including the ‘constitutional avoidance canon’, a non-redundant subset of the ‘last resort rule’ (Kloppenberg 1994). This canon is well articulated in United States vs. Coombs, a decision from 1838 in which the court said:

If a section of an act of Congress admits of two interpretations, one of which brings it within and the other presses it beyond the constitutional authority of Congress, it is the duty of the Supreme Court to adopt the former construction, because a presumption never ought to be indulged that Congress meant to exercise or usurp any unconstitutional authority …

The canon has repeatedly been employed, including in such early cases as Parsons (1830), Grenada County (1884), Hooper (1895), Knights Templars (1902), United States v. Delaware & Hudson (1909; Linares-Quintana 1998), Panama (1924) and Crowell (1932). By the first decades of the twentieth century, it was a settled canon. Critics disagree about whether it aims to prevent the court from deciding constitutional questions or from considering them (e.g. Katyal and Schmidt 2015; Nelson 2015).

The constitutional avoidance canon is indistinguishable from one way of conceiving what is often known as ‘interpretation in conformity with the constitution’, a doctrine born in Germany (verfassungskonforme Auslegun; Alexy et al. 2019) that has found reception in Latin America. Under this version, both the canon and the doctrine are predicated upon the deference owed to the rule’s author – most especially legislators. By choosing a reasonable interpretative option that is in line with the constitution, the courts avoid offering an explicit constitutional decision, particularly one striking down the challenged rule. They save the rule ¹² by presupposing that the authors did not intend to violate the constitution. But there is another way of understanding ‘interpretation in conformity’. It does not centrally involve a question of deference but a mandate to choose an interpretation that conforms with the constitution or relevant treaty, even if it is abundantly clear that the rule’s authors intended otherwise. ¹³ To the extent that deference is implicated, it is mainly due to the constitution or treaty, and not so much to the rule’s author. In this version, the constitutional question is not avoided but rather tackled head-on. Colloquially, this can lead to a partial declaration of unconstitutionality – since the norm that the rule’s authors intended to enact is partially quashed – and can apply regarding any legal rule. ¹⁴

The Ashwander rules are jointly known as the avoidance canons, and they connote the idea that, if possible, constitutional review must be dodged whatever the state action being challenged – the language in Rescue Army. When Schauer (1995) writes about them, he says that ‘We accept the authority of the courts to invalidate the decisions of other bodies, but just barely.’ Yet, as described, the court routinely reserves the stronger language concerning the ‘gravity’ and ‘delicacy’ of judicial review and like expressions to challenges to statutes. As Schauer (1995: 71) says, ‘The judicial invalidation of an Act of Congress, and to a lesser extent the invalidation on constitutional grounds of state legislation or the acts of federal and state executives, is disfavored in American constitutionalism.’ These ideas appeared in the work of early scholars such as Gustav Endlich (1888: 246), who said, ‘A presumption of much importance in this country … is that a legislative intent to violate the constitution is never to be assumed, if the language of the statute can be satisfied by a contrary construction.’

Criticism has been levelled against these canons (e.g. Katyal and Schmidt 2015; Kelley 2001; Marshall 1990; Schauer 1995). Judging them – or, more generally, analysing the manifold justifications and criticisms of judicial review – far exceeds the scope of this article. However appropriate, the rules or canons are intelligible. Taken jointly, they do not apply with the same force to rules that are not statutes, and some of them – such as the constitutional avoidance canon – seem to apply only to statutes. Non-trivial implications often stem from the invalidation of other rules, including the risk of legal instability and the sacrifice of expectations. But weighty considerations that are present when it comes to democratic legislation are absent or weakened in those cases.

The presumption of constitutionality

As denoted by Endlich’s (1888) words, the constitutional avoidance canon is tied to the US court’s acceptance of the ‘presumption of constitutionality’ of statutes (hereinafter the ‘presumption’). The central notion is that ‘doubts’ involving a challenged rule are resolved in the rule’s favour. Regardless of whether this is truly a presumption or what kind of presumption it is (Franck 2019; Roberts 1959), it is related to the canon as a measure of deference. Still, its meaning is not settled, since one can identify three (perhaps four) versions of it in both court opinions and the literature.

Although in some cases it can become blurry, a key conceptual distinction is between, on the one hand, doubts about the interpretation of the constitutional provisions at stake in a case, and, on the other, doubts about a challenged rule (Ferreres Comella 2007). The word ‘doubt’, however, does not exhaust the presumption’s meaning, and the link with evidentiary matters can be misleading. In some instances, the issue is whether there are reasonable interpretative alternatives to preserve a rule. The following versions differ with regard to both their strength and the source of ‘doubts’. Now, a presumption assertion does not have a clear place if a court does not have doubts about the rule’s constitutionality. Like an old contribution says, ‘The suspicion sometimes arises that the grandiose phrase “all is to be presumed in favor of the legislative action” is in fact meaningless’, since one may conclude that, if the decision is to uphold the statute, it is employed ‘merely as a rationalization of the result reached’ (Columbia Law Review 1931: 1137).

A first version of the presumption – the strongest – was well formulated by Justice Washington in 1827. It applies to doubts about the constitution, the challenged rule or both. Washington said:

If I could rest my opinion in favor of the constitutionality of the law on which the question arises on no other ground than this doubt so felt and acknowledged, that alone would in my estimation be a satisfactory vindication of it … It is but a decent respect due to the wisdom, the integrity, and the patriotism of the legislative body … to presume in favor of its validity until its violation of the Constitution is proved beyond all reasonable doubt’ (Ogden 1827).

In Harris (1883), the court said that, ‘Proper respect for a coordinate branch of the government requires the courts … to give effect to the presumption that Congress will pass no act not within its constitutional power.’ This presumption will prevail ‘unless the lack of constitutional authority to pass an act … is clearly demonstrated’ (Harris 1883; Metzger and Morrison 2013: 1729–31; Grenada County 1884).

In Ashwander, Brandeis quoted this paragraph by Chief Justice Waite in Sinking Fund Cases (1878):

[T]his declaration [of unconstitutionality] should never be made except in a clear case. Every possible presumption is in favor of the validity of a statute, and this continues until the contrary is shown beyond a rational doubt. One branch of the government cannot encroach on the domain of another without danger. The safety of our institutions depends in no small degree on a strict observance of this salutary rule.

These ideas – which can be analyzed regardless of whether the court applied them consistently – resonate with the work of James Bradley Thayer (1893). Thayer is generally associated with the deferential ‘rational basis review’ in the United States (Ferreres Comella 2007), but his ideas cut deeper, applying to doubts about the constitution itself. Thayer wrote that the primary, and in most cases, final determination of the constitutionality of a federal statute was ‘intrust[ed] to the legislature’ unless it was shown that ‘those who have the right to make laws have not merely made a mistake, but have made a very clear one, – so clear that it is not open to rational question’ (Thayer 1893; see also Franck 2019; García-Mansilla 2013: 17–18; Whittington 2019: 5). Constitutional meaning often offered a ‘range of choice and judgment’ and judges should defer to the legislature’s choice even if this entailed putting aside what they considered the most proper reading (Franck 2019; Thayer 1893). For Thayer, ‘whatever choice is rational is constitutional’ (Franck 2019; Sunstein 2022; Thayer 1893). The judicial function did not involve ascertaining the true meaning of the Constitution but rather deciding whether the legislature’s reading was unreasonable (Franck 2019). A federal statute could only be struck down in ‘very clear’ cases of constitutional error. Thayer’s main expertise was evidence, and he acknowledged that this influenced his language (Franck 2019). He offered as an example of his view the judicial review of juries’ verdicts, which could only be overturned if ‘reasonable men could not fairly find as the jury have done’ (Ferreres Comella 2007; Thayer 1893; White 1993: 77).

Thayer invoked in his favour a tradition of judicial decision-making. He cited Ogden and Sinking Fund Cases, both mentioned above, as well as a Pennsylvania Supreme Court decision from 1811 (Franck 2019; Thayer 1893: 140). He also quoted Thomas Cooley, who in 1871 had written that ‘the power to declare a legislative enactment void is one which the judge … will shrink from exercising in any case where he can conscientiously and with due regard to duty and official oath decline the responsibility’ (Cooley 1871: 159). Cooley added:

The legislative and judicial are co-ordinate departments of the government, of equal dignity; each is alike supreme in the exercise of its proper functions, and cannot directly or indirectly, while acting within the limits of its authority, be subjected to the control or supervision of the other, without an unwarrantable assumption by that other of power which, by the constitution, is not conferred upon it (Cooley 1871: 159).

In sum, courts would ‘never declare a statute void, unless the nullity and invalidity of the act are placed, in their judgment, beyond reasonable doubt’, and a ‘reasonable doubt must be solved in favor of the legislative action, and the act be sustained’ (Cooley 1871: 182). In this way, the presumption ‘creates a canon of construction’, as Magnet (1980: 116) has pointed out in the Canadian context.

To a degree, the US court preserved this language, although during the interregnum of the Lochner era it did not show itself ‘disposed to heed the Thayerian call for judicial deference to legislatures’ (Whittington 2019: 147). Since the middle of the twentieth century, however, it was relaxed due to a change in the underlying justification. In cases such as Cooper (1958), the court reclaimed ‘the ultimate authority to interpret the constitution’ (Hessick 2010: 1456). Thus, one observes in the court’s subsequent decisions a second version of the presumption that is weaker than the first in two senses. First, it no longer refers to doubts about the Constitution. Second, it is conditional upon the subject. As Metzger and Morrison (2013: 1729–30) write, the notion is that the court will not abdicate its mission to say what the constitution means and demands – only that it will not ‘lightly arrive’ at a determination of unconstitutionality and will reserve it to clear and unavoidable showings of it. ‘Ambiguity in statutory meaning’, the condition behind the avoidance canon, ‘can be one reason why a statute might not be clearly unconstitutional’ (2013: 1731). The court would also recognize the legislature’s ‘superior institutional ability to make the factual determinations upon which an enactment’s constitutionality might depend’ (2013: 1732). Yet the presumption is supposed to be tempered when ‘fundamental rights are implicated’ (2013: 1729). The more recent and limited Chevron doctrine (1984), which implies deference to administrative agencies when the statute they administer is ‘silent or ambiguous with respect to the specific issue’, and provided their interpretation is reasonable, might also fall into this version. The rationale behind this doctrine is based both on the alleged technical expertise of agencies and the accountability of the latter to the president (Sunstein 2022).

A third version more narrowly involves factual deference. The court asks ‘whether there is a conceivable set of facts that would justify the law’ given the court’s interpretation of the constitution (Franck 2019; Hessick 2010: 1460). So they will ‘assume facts necessary to satisfy the constitutional test under which the legislation is being evaluated’, even if they do not exist (Hessick 2010: 1452). It will suffice that ‘a rational legislator could have reasonably thought that they exist’ (Hessick 2010: 1453, emphasis in original). This version is inextricable from tests evaluating the rationality of legislation, and is related to the ‘rational connection’ step in the proportionality review. ¹⁵

Finally, one might conceptualize a fourth, weaker version according to which a rule is in force until struck down if it has ostensibly been enacted following the proper procedures. In a system of concrete review, this version entails the requirement that a party challenging a rule shows how it is harmed by it. While this requirement is part and parcel of the other versions, here it consists merely of a threshold – a party’s challenge must be sufficiently serious. Importantly, this procedural threshold does not speak of a special disposition of courts concerning the rule’s content, which is why it is not in the same camp as the other versions.

In the United States, the first two versions chiefly apply only to statutes – except, in the case of the second, for the space reserved to an agency in charge of administering a statute that is silent or ambiguous with respect to a particular issue. The fourth (to the extent that it is considered) applies to all rules adopted by officials serving under colour of title (Constantineau 1910). But the Court seems largely to have extended the third to executive and administrative action as well. A year before Ashwander, in Pacific States Box (1935), Justice Brandeis writing for the court said that the ‘rebuttable presumption of the existence of a state of facts sufficient to justify the exertion of the police power’ applies to any such exertion. The sparse justification given for this extension was that those exertions were an ‘exercise of delegated power’ (Pacific States Box 1935; Ponomarenko 2018: 1419).

The main notion appeared to be that courts reviewed the constitutionality of delegations – particularly of legislative delegations – themselves, rather than the ‘substance of the decisions reached’ by administrative agencies (Ponomarenko 2018: 1420). Ponomarenko shows that this rationale was short-lived, but argues that the presumption lived on, based on what the author largely sees as an improper extension to the administration of claims made in the context of legislative action. These include ‘authority-based’ claims concerning the legal or political authority of administrative agencies, arguments about the respect for federalism or the epistemic potential of administrative procedures, and doctrinal and interpretative considerations (Ponomarenko 2018: 1423–48). According to the author, none of these is sufficient to justify the deferential approach taken by the court. Epistemic arguments are perhaps the strongest, but they are irrelevant when the presumption is announced, since this is often done precisely when the agency’s factual considerations are deficient or lacking (Ponomarenko 2018: 1439).

In any event, while in Rescue Army (1947) Justice Rutledge referred to constitutional review as ‘the most important and the most delicate of the Court’s functions’, it is in its application to legislation that the court has uttered its strongest language concerning deference.

The language of deference

In Ashwander (1936; Kloppenberg 1994), Brandeis wrote that the court ‘has frequently called attention to the “great gravity and delicacy” of its function in passing upon the validity of an act of Congress’. He cited several decisions, including Ex parte Garland from 1866, in which the court said:

It is at all times the exercise of an extremely delicate power for this court to declare that the Congress of the nation, or the legislative body of a State, has assumed an authority not belonging to it … In the case of an act of Congress, which expresses the sense of the members of a coordinate department of the government, as much bound by their oath of office as we are to respect that Constitution, and whose duty it is, as much as it is ours, to be careful that no statute is passed in violation of it, the incompatibility of the act with the Constitution should be so clear as to leave little reason for doubt before we pronounce it to be invalid.

Three years later, in Hepburn (1869), the court wrote that ‘the delicacy and importance of this question has not been overstated in the argument’, adding that it ‘always approaches the consideration of questions of this nature reluctantly’ and that ‘acts of Congress must be regarded as constitutional unless clearly shown to be otherwise’. In Adkins (1923), a case from the Lochner era, the court defined that duty as ‘one of great gravity and delicacy’. Later, in Blodgett (1927), concurring Justice Holmes wrote:

Although research has shown and practice has established the futility of the charge that it was a usurpation when this Court undertook to declare an Act of Congress unconstitutional … to do so is the gravest and most delicate duty that this Court is called on to perform.

While this language is partly a reflection of the long-relaxed first version (resonant of Thayer’s criterion), it has survived, signalling the court’s prioritization of statutes. The court has employed the ‘gravest and most delicate duty’ phrase countless times to date either in majority, plurality, concurring or dissenting opinions (e.g. Fullilove 1980; Northwest 2009; Rostker 1981; Rust 1991).

The primacy of statutes is predicated upon straightforward reasons, as noted above. Paired with the fact that judges usually lack democratic credentials of their own and are largely unaccountable, those reasons have famously led some scholars to distrust strong judicial review (e.g. Tushnet 2009; Waldron 2006). Jointly, these factors also provide the backdrop to the development of judicial canons of self-restraint. While these canons rest on the idea that there are legitimate exercises of judicial review, they call attention to valid concerns about the task. Even defenders of judicial review, including Bickel (1986), often have these concerns in mind. Bickel strove to keep judicial review within a sphere that was respectful of the separation of powers, and he advocated for courts to embrace ‘the passive virtues’, which meant accepting those canons. The invalidation of unilateral executive action is sprinkled through his text, but Bickel was especially worried about the striking down of legislation.

Executive orders in a presidential system are not devoid of legitimacy since they rely upon the electoral credentials of the president, and a degree of deference must be owed to them – particularly when they regulate matters within the executive branch, a point discussed below. Yet the executive only weakly represents the pluralism and disagreement that exist in a society. Also, checks and balances are germane in this case since the executive potentially can infringe upon Congress. Such a vociferous critic of judicial review of legislation as Waldron (2006) has endorsed the review of executive action. According to Waldron (2006: 1353–54), the reasons against the former do not weigh against the latter, since the executive is ‘subject to the principle of the rule of law’ (see also García-Mansilla 2013: 82). From executive orders downwards, the legitimacy of a rule diminishes as political accountability decreases. Both the general expertise that some administrative offices may possess and the participation they may afford citizens and stakeholders do not quite offset their lack of political accountability and representativeness.

Even if, regarding factual considerations and, in some cases, the interpretation of silent or ambiguous statutes, the US court has extended the administration the deferential treatment it has given the legislature, the strongest language revealing self-restraint was originally – and to a large extent still is – reserved for statutes. As will be seen next, while the Supreme Court of Argentina applied most of the previous canons and presumptions, it eventually changed their logic. The court would stretch the ‘great gravity and delicacy’ rationale, and at times the strongest version of the presumption, to cover any rule without any consideration to its pedigree. In doing so, it kept itself within the ostensible logic and language of familiar doctrines while either distorting them or hollowing them out.

IV. Self-restraint in Argentina

The Argentine Constitution, including its federal judiciary provisions, was modeled after the US Constitution. Since the Supreme Court began its work in the absence of any discernible domestic case law some 70 years after its North American counterpart, it is unsurprising that it fixated on the latter.

The influence of US case law

Early on, the court relied on the US Supreme Court case law, especially in discussing its jurisdiction (Miller 1997). Miller cites the hyperbolic declaration of a US Ambassador to Argentina who in 1891 said, ‘No leading lawyer here is without his complete set of our U.S. Supreme Court reports’ (Miller 1997: 1544). The Argentine Court would often go as far as to declare that US constitutional case law was binding on it (Miller 1997: 1561; Sojo 1887). This is a point that Domingo F Sarmiento, an influential writer, polemicist and politician who was president from 1868 until 1874 made in his commentary on the new Argentine Constitution (Miller 1997: 1525–26). The court’s first decisions on the pertinence of judicial review of legislation from the late nineteenth century bear a remarkable resemblance to those in the United States, including a discussion, as in Marbury, of the legislative power to extend the court’s original jurisdiction (Sojo 1887; Miller 1997: 1547). ¹⁶ While the US court’s influence began to wane ‘beginning in the late 1890s’ (Miller 1997: 1546), the Argentine court would for decades cite its counterpart from time to time. As late as 1963, the court said that Argentina had a model of judicial review borrowed from the United States (Abalos 1963).

Among the full panoply of imported self-restraint doctrines are the Ashwander rules as well as (some version of) the presumption of constitutionality, at times with evocations of the strong Thayerian language. Even so, in good measure because of institutional breakdown and the corresponding judicial legitimation of it, the court distorted those doctrines by applying them outside their obvious scope conditions without much – or any – rationale to anchor them. While the court did not always apply these doctrines and, even when cited, they did not always play a tangible role, their use reveals an objectionable degree of subterfuge and the potentially normalizing power of judicial discourse.

Avoidance canons and the presumption of constitutionality

In 1874, over a decade before it struck one down for the first time, the Argentine Court said a statute could be invalidated if it was ‘absolutely incompatible’ with the Constitution (Avegno 1874; Linares 1989: 136–37). This canon stemmed from ‘the respect owed the high Powers that concur in the formation of statutes’. In 1909, the court already referred to the ‘well-settled principle’ according to which the judiciary will not repeal a statute barring a ‘clear and undisputed opposition’ between it and the Constitution (Banco Hipotecario 1909). In a handful of cases invoking this notion, the court cited the US Supreme Court, including in a tax law case under the first Perón administration, where it said that ‘any doubt should be resolved in favor of the statute’s validity’ (La Franco 1948).

In 1960, the Court quoted the US Court, together with its own case law, to the effect that ‘no higher duty rests upon this court than to enforce, by its decrees, the will of the legislative department of the Government … unless such statute be plainly and unmistakably in violation of the Constitution’ (Cine Callao 1960; Northern Securities 1904; García-Mansilla 2013: 21). With some Ashwander connotations, these phrases seem to approach the strongest version of the presumption described above, which the US Supreme Court originally embraced but relaxed afterwards. This, of course, is independent from the court’s actual motivation in employing this language – in some cases, it may have been that the court wished to uphold the statute under challenge all along without entertaining any doubts.

The 1874 decision quoted in its support Alexander Hamilton’s ‘evident opposition’ phrase concerning statutes from Federalist #81, a language that resonates with Thayer’s thinking (Franck 2019: 411; Hamburger 2008: 312–13). In 1958, the court quoted both that decision and Justice Marshall (wrongly identifying the source) in Fletcher v. Peck (1810). Reviewing the constitutionality of a law (a provincial statute) was a matter of ‘much delicacy’, and to strike it down, the opposition between the law and the constitution ‘[should be such that the judge feels a] clear and strong conviction of [their] incompatibility’, so any doubt must be resolved in favour of the law’s validity (Degó 1958; bracketed text in original decision; Linares-Quintana 1998).

Together with case and controversy, ripeness and standing requirements, and the ‘last resort’ rule, the court announced the constitutional avoidance canon concerning ‘doubts’ about the challenged rule. For example, in 1924 it wrote that a statute’s construction showing it in harmony with the Constitution must be preferred, under ‘the presumption that Congress has striven to keep itself within its constitutional powers’ (Empresa de los Ferrocarriles 1924). The court would employ the canon several times in the ensuing decades (e.g. Carniol 1947) until the present. In some decisions announcing the canon, it first used language close to the first version of the presumption, adding next that the latter was complemented by the notion ‘that laws must be interpreted, inasmuch as it is possible to do so without violence, in the manner most in line with the Constitution’ (Servicios Aéreos 1953).

In parallel, especially in the first decades of the twentieth century as the state began to intervene more decisively in the economy and, above all, with the first brushes with institutional breakdown, the court began to extend a more general presumption of constitutionality to the acts of public authorities, regardless of their origin (Linares 1989: 213). ¹⁷ This fusion in the court’s reasoning of statutes and other regulations would be accentuated in the ensuing decades. Under dictatorship, it would assist in the validation of de facto rules. Under democracy, it would often strengthen executive authority in Argentina’s presidential (or hyper-presidential) system. In 1932, for example, two years after the first coup, the court wrote that a ‘presumption of constitutional validity’ attached to both statutes and executive orders – involved in the case was an order issued by the military administration – which meant for the court that a tax was mandatory until it was struck down (Chiappe 1932). In future decisions, the court would adopt a stronger version of the presumption.

In 1944, the court penned language resembling the statutory avoidance canon in another case concerning an executive order issued by the military (Lacour 1944). It said that, ‘in the interpretation of statutes [leyes], a construction must be preferred that best concords with the guarantees, principles, and rights established in the Constitution’ and that ‘only if the contradiction is evident and no other … interpretative alternative exists’ will the court choose it and strike down the rule. In its support, the court cited both its previous case law and the US Supreme Court decision in Crowell, although both groups of sources announced deference to the legislature. By the mid-1940s, the reference to the statutory source of the avoidance canon had all but vanished from the court’s record, in conjunction with its equation of rules adopted by the military with democratic statutes described in Part II. The court occasionally cited both the canon and other Ashwander-like doctrines as part of the reasoning to dodge constitutional questions and defer to the authors of rules of various pedigrees. The rationale in some of these decisions would include language closely resembling the first version of the presumption. In 1959, a judgment involving a rule passed by a dictatorship cited case law concerning statutes to require the ‘absolute incompatibility’ between the challenged rule and the constitution as a condition of a strike-down (Federación de Obreros 1959). While in some cases the court included a reference in passing to the hierarchical condition of statutes (e.g. Aserradero 1961), it largely extended the same treatment to any regulation.

This blending of rules would become routine. In a previously mentioned decision from 1960 employing expressions closely resembling the Thayerian language to address a statutory challenge (Cine Callao; Linares-Quintana 1998), the court ‘especially’ cited the precedent from 1959 discussed in the preceding paragraph involving a military order. And, in a decision announced in 1973 under a dictatorship concerning such an order, the court cited a previous judgment saying that in reviewing the constitutionality of a rule in that case, an actual statute, the court had to act with the utmost circumspection, respecting the other branches’ exclusive powers (Puerta 1973).

Although largely explained as the product of institutional instability, this trend is puzzling. While one may find prudential considerations for the court’s deferential treatment of the rules passed by military authorities, there is hardly any normative justification of it (e.g. Nino 1985: 89–107). Under democracy, a degree of judicial restraint is surely merited concerning regulations issued by the executive or the administration within the purview of their authority. This includes the requirement that a constitutional question not be ‘passed upon’ unnecessarily, especially regarding so-called autonomous decrees (decretos autónomos), those executive orders issued by the president to manage the administration, which in Argentina is organized within the executive branch. But the level and justification of such restraint can never be as strong or as indiscriminate as in the case of statutes. To the general reasons provided above, one can add that administrative law in Argentina has largely been exempt from legislative oversight. It mostly responds to the federal or state (provincial) executives and its procedures have not been remarkably transparent or participatory. Furthermore, some executive action in such a strong presidential system as Argentina’s can potentially sidestep the authority of Congress. Apart from the regulatory orders necessary for the execution of statutes, this includes executive orders and administrative action concerning powers delegated by Congress and emergency orders (decretos de necesidad y urgencia). The latter have been used frequently since the 1990s, partly to address repeated economic crises. ¹⁸

In sum, a melting pot was brewing hot in the court’s decision-making, fusing statutes with any other regulations without much justification.

V. The slogan

An expression the court used in 1961 became an occasional catchphrase to be included at the beginning or the conclusion of an argument of self-restraint. Since this phrase’s usage problematically combined and confounded most notions of deference and self-restraint under vastly different circumstances, the remainder of this article is centred around it. Striking down a ‘statute’ (ley), said the court on that decision, was ‘an act of extreme institutional gravity and must be considered the last resort [ultima ratio] of the legal system’ (Rasspe 1961; García-Mansilla 2013: 15). The slogan denotes two interconnected notions. First, a rule’s invalidation must be reserved for eminently clear and unavoidable cases, and this is so, second, because of the extreme gravity involved in those declarations. The court’s language ostensibly resembled the Rescue Army phrasing relating to the ‘last resort rule’ but, more appropriately, it echoed Justice Holmes’ stronger words in Blodgett (1927).

The 1961 case was decided during a very brief spell of electoral rule. The ‘statute’ in question was an order passed by a military government concerning the confiscation of enemy property at the end of World War II, when Argentina belatedly declared war on the Axis powers. The court inserted the slogan while dismissing the challenge, adding that the plaintiff had both consented to the order and waited years to defy it, thus failing to show how the decree injured him. This approaches the threshold requirement. The court would later complement the phrase by noting that striking down a rule is ‘the most delicate power vested upon a court’ (e.g. Arancibia 1980). Remarkably, the court most often penned the ‘extreme institutional gravity, last resort’ language as a freestanding slogan without elaboration. It simply relied on the power of a familiar-sounding phrase.

Consequently, both the Ashwander doctrines and the language reserved to statutes – a stronger version of the presumption, the constitutional avoidance canon and the considerations concerning the singular gravity and delicacy of a strike-down – would be mechanically invoked in consideration to any rule, including those passed by the de facto military administrations. The slogan’s strong language was used more often after the 1976 coup and would persist until the present. In the 60 years between 1961 and 2020, the court inserted the ‘extreme institutional gravity, last resort’ slogan (or an almost indistinguishable language) to reject a petitioner’s challenge around 150 times, including almost 20 cases where a petition was granted on other grounds. In around one-third of those total cases, as is otherwise not uncommon, the court referred for brevity’s sake to the opinion written by the Procurador General, the head of the federal judiciary’s prosecution office who issues a non-binding opinion before the court’s opinion.

Taxonomy

The slogan seemed to accompany or encapsulate several ideas concerning self-restraint, although it is often hard to tell them apart. It is unusual for the court to describe in full the challenge to a rule, and its reasoning is typically terse.

1. 1. In some cases, the court applied the ‘last resort’ slogan in a manner similar to the general doctrine that bears that name in the United States, saying that striking down a regulation is a measure to be taken only when ‘a strict necessity requires it’ and not when a decision can be found on non-constitutional grounds, although the court is in principle barred from interpreting provincial or ordinary law (e.g. Banco Europeo 1983; Nuñez 2012; Linares 1961: 92–98).

2. 2. A few times, the court nominally approached the statutory avoidance canon, expressing the idea that a rule – any rule – must not be struck down if a construction of it – including a systematic interpretation of the rule in its connection with others – shows it is in line with the constitution (e.g. Banco de Río Negro 1981; Llerena 2005).

3. 3. Most of the time, the court invoked the slogan as part of an argument about either the lack of standing to challenge the rule or the challenge’s mootness or frivolousness. The court considered either that the challenger had not provided convincing arguments to show that the rule was invalid or had injured them, or that the challenge was too generic or succinct (e.g. Cosimano 1977; Bravo 1978; Forstmann 1979; Bigongiari 1979; Paredes 1980; Philco 1984; Furbia 2005; Andrada 2005).

4. 4. On several occasions, the court cited the slogan as the introduction to, or conclusion of, an argument upholding the rule (e.g. Halladjian 1980; Greco 1982; Bruno Hnos 1992; Valmor 2006). Sometimes it accompanied this with the long-used statement that it was not incumbent upon it to replace the judgment of the official issuing the rule or to analyze the rule’s wisdom (e.g. Disco 1989).

5. 5. Finally, in some cases the court invoked the slogan within the context of a challenge to express the seriousness of the task, which required careful and detailed consideration of the issue (Mexicana 2008). This does not add much since a court must always take the parties’ arguments seriously.

The indiscriminate language of deference in all cases is unjustified since it wrongly conveys the twin notions that the author of any rule similarly deserves the highest level of respect in terms of judicial scrutiny and that any decision striking down a rule is equally serious. Regarding point (2), it must be remembered that both the presumption of constitutionality and the strongest language of deference were largely reserved for statutes in the US model. While, if possible, all rules must be made in line with the constitution, it is the legislature (subject to exceptions) which deserves the highest consideration in the process. Use (point 4) is uncalled for in cases where the court concludes that the rule is valid under both its interpretation of the constitution and the rule’s most straightforward reading. In some decisions, the court does seem to use the language in this way. In others, the court conveys a sense that it is only analysing the reasonableness of the regulation. This may be so either because it embraces a strong Thayerian stance – namely, that the officials’ constitutional interpretation is valid if reasonable even if it diverges from that of the Court – or, perhaps more likely, because it commits to rational basis review, namely that there is nothing unreasonable in the officials’ selection of ends and means even if the regulation is not grounded on careful factual considerations. The court’s formulaic language does not always allow telling these uses apart.

Under one interpretation, use (point 3) is close to the weakest version of the presumption, which also does not justify the strong language of deference. The court obviously does well in rejecting frivolous challenges, although if all the work the slogan does is to add to the idea that a challenger must prove their allegation, it is a commonsensical argument that applies to many claims in front of a judge. Critically, however, point 3 is in several cases indistinguishable from the rest. For when it considers a challenge lacking, the court at times seems to imply that it is not convinced that the rule is clearly unconstitutional, something that reaches beyond the minimum procedural threshold. And this can be either because it considers the rule valid tout court or because of interpretative doubts, which it resolves in the challenged rule’s favour. The court’s bare language is again not helpful in this regard.

To sum up, the phrasing of the ‘last resort rule’ in the United States, weakly tied to a presumption and originally indicating that a declaration would not be made if unnecessary, was transformed to reinforce the presumption that all legal rules comply with the Constitution and as a recipe for self-restraint given the allegedly gravest implications following any invalidation. A statute was considered institutionally central to the nation’s life as an arcane agency regulation, or an order implemented by the military. In its mechanical application across the board, the slogan became a catch-all phrase.

The slogan under dictatorship

Although it continued to do so sporadically, the court invoked the slogan more often after the 1976 military coup – over four times a year in the seven and a half years the murderous dictatorship lasted (1976–83). In around a dozen cases, the dictatorship court used the slogan in challenges to statutes – national or provincial ones – usually in connection with alleged insufficiencies in the challenger’s case or as part of an argument to the effect that the rule was constitutional (uses – point 3 and point 4 above). This is the most persuasive invocation of it, one made by a court serving under, and appointed by, a dictatorship addressing a statute from a democratic legislature.

Still, most uses (around 20) referred to challenges to rules passed by the military. Right after the coup, the authorities decreed that what the previous democratic government, following French law, had labelled ‘decree-law’ (decretos-leyes), namely those general rules passed by a dictatorship, were now to be simply called ‘statutes’ (leyes) (Diana 2012: 44–45; Gordillo & Campolieti 2013). Yacovino (2013: 40) rightly points out that this change was meant to ‘erase the authoritarian connotations’ associated with the rules’ origin (see also Diana 2012: 44–45). Although this legal chicanery did not authorize it to attach any type of presumption to these regulations, the court did exactly that. For example, in 1976 it repeated the slogan right before concluding that such a rule was constitutional, referring to it as a ‘statute’ (ley) (Baldini 1976). In some cases, the court even identified the rules issued by the military as emanating from the legislative branch (e.g. Esteban 1980).

The court occasionally used the slogan together with a statement to the effect that rules passed ‘by the sitting authorities are presumed valid’. It did the latter in a decision from 1981 dismissing a summary challenge to a set of rules enacted by the military governments. The Jehovah’s Witnesses denomination was prevented from being registered as a religion, a requisite for its seminarians to be exempted from military service (Falcon 1981). The court said that only the rule’s ‘manifest and irreconcilable’ incongruence with the Constitution could lead to its invalidation in a summary proceeding, a language with Thayerian overtones.

In 1979–80, the court invoked the slogan as one of the considerations to reject challenges from plaintiffs who had been fired because of a dictatorship’s rule against strikes (Bigongiari 1979; Esteban 1980). In the latter judgment, the court said that the right to strike had been suspended by the state of siege in place in the country and that this validated the rule. In both, it conveyed the notion that the rule had not been shown to be unconstitutional (although it did not bother to fully detail the challenges), and thus could not be struck down because of the extreme institutional gravity that would follow such a conclusion. In 1980, it dismissed a challenge to a preposterous rule introduced by a military administration concerning personal names. A parent wanted to name his daughter ‘Anouche’, but the rule banned ‘foreign’ names. The court invoked the slogan and resorted to a favorite motto – no right is absolute—to conclude that the rule’s chauvinistic criterion was not unreasonable; it said that a nation’s language was connected to its ‘spiritual identity’ (Halladjian 1980).

In a case from 1983 featuring ominous undertones given the illegal appropriation of hundreds of children whose mothers were murdered or disappeared by the dictatorship, the court cited the slogan to dismiss a challenge for containing generic allegations (Petit Petit 1983). This exemplifies the complex nature of seemingly procedural or threshold decisions. A mother complained that, despite her opposition and the lack of a decision depriving her of custody over her daughter – something the existing rule enacted by a previous dictatorship did not require and that she challenged – a court had given the latter in adoption. The court said that the lower court had considered the mother’s condition and decided in consequence. If losing a child is a generic complaint, anything is.

As noted, in invoking the slogan, the court under the dictatorship was implementing a transmutation. ¹⁹ It was superimposing an image of normalcy upon the abnormal, applying tried and familiar constitutional canons outside their straightforward scope conditions and without elaboration. The court was reclaiming an entire constitutional tradition to support the rules, at times masking their illegality behind the language of liberal and democratic constitutionalism. In this and other ways, the court’s justices were contributing to lending legitimacy to an illegal regime. Once the modified doctrines had been applied time and time again and were intertwined in cross-citations with the application of similar notions, they were sufficiently settled and assimilated for future courts to keep them, even though a revision and a refined understanding were in order.

The slogan under democracy

The Supreme Court from 1983 took considerable distance in several respects from its immediate past. This includes its judgments subordinating the military rules’ validity to Congress’s (explicit or implicit) recognition and, more generally, its decision-making in cases involving freedoms, in some cases reproaching the dictatorship’s repressiveness and perfectionism (e.g. Justice Petracchi’s famous opinions in Fiorentino 1984 and Bazterrica 1986, and Justices Petracchi’s and Belluscio’s opinions in Arenzón 1984). While the Menem Court (1990–2003) represented a setback in some of these aspects, both the 1994 constitutional amendment and subsequent personnel changes had the effect of strengthening the judicial protection of individual and social rights. As is healthy for an institution acting in a democracy, the court is routinely criticized – sometimes (as lately) harshly – but it is a crucial institution in today’s Argentina.

Still, even if they are subtle, there are still connections with the authoritarian past. This is not to suggest that such connections undermine the court’s mission, since they do not. But the connections matter for at least two reasons. One is that, as discussed above, it shows the inertia that the legal manifestations of that past may have once they become entrenched in discourse, which only a steadfast reckoning may overcome. The second reason is that courts, and particularly one in such a need for legitimization as the Argentine Court, must avoid artifice. This invites the court to offer public and well-articulated reasons for the doctrines to which it resorts, whether they are the bundle of self-restraint doctrines which it uses sporadically or others.

Concerning this article’s topic, the court under democracy roughly has used the slogan as before, albeit in a somewhat less problematic way. From 1984 to 2020, it employed it just over 100 times in dismissals to challenges, an average of three times a year. In slightly over half of the cases, the court applied both the slogan and concomitant considerations to statutes, including those dealing with the country’s legendary economic emergencies. It did so usually to dismiss challenges or to uphold the rule, in some cases selecting an alternative construction. In a few decisions, it seemed to allude to the statutory nature of the challenged rule, adding that only if the rule was clearly unreasonable or the violation was manifest would it strike it down (e.g. Lapadu 2004; Ministerio de Cultura 2008; Ministerio de Relaciones Exteriores 2016; González 2017). ²⁰ In this, it preserved something of the Thayerian tradition it had announced previously. Yet, over the years, the court softened this stance partly by introducing the notion of suspect categories to deal with discrimination (in particular, that based on national origin), which has led it to presume the unconstitutionality of rules touching upon those categories unless a state interest was found compelling (Calvo y Pesini 1998; Hooft 2004; Garay 1989). It has also sometimes relaxed the presumption concerning the infringement of other fundamental rights. In one case, the court used the slogan to dismiss a challenge to a statute that had annulled the appointment of university professors made under the dictatorship, underlining the irregular nature of such appointments by an illegitimate regime (Gamberale 1989).

Although still unfolding, these developments are steps in the right direction for a democratic society that recognizes the centrality of statutes and the need for a degree of judicial deference towards the legislature, while also being fully committed to the protection of rights. They still have some shortcomings, not least in terms of consistency. One example where the court failed to relax the presumption in the presence of fundamental rights is from 2018 (Asociación Francesa 2018). A hospital was found responsible for malpractice in the birth of a girl with cerebral palsy and went bankrupt. The child’s parents asked to be paid with priority given their predicament, which meant changing a statute’s ranking of payments (González-Bertomeu & Álvarez-Ugarte 2019: 20). The court dismissed the claim. Citing the slogan in the opinion, it found the statute reasonable. It said that defining such a ranking was ‘the exclusive power of the other branches’. The dissenting justices, however, pointed to the multiple constitutional and treaty rights recognized in relation to children and the obligations the state had assumed in this area.

At any rate, this overall salutary use of the slogan and underlying doctrines of self-restraint represents only part of the picture. For the court has applied the slogan in about 50 decisions to regulations that were not statutes. In slightly over fifteen, it did so with respect to the dictatorship’s rules (e.g. Industria Delta 1984; Bruno Hnos 1992; Prado 2012), including a secret rule structuring intelligence activity (Giardelli 2002). Incidentally, in 1988, the court summarily dismissed another challenge to the repressive rule concerning personal names (the name was now ‘Tracy’) by referring to the dictatorship case above that cited the slogan (Ofmann 1988); Justice Fayt filed a dissent. In another ‘slogan’ decision that found Fayt dissenting, he underscored the ‘spurious’ nature of the challenged regulation, a criminal rule that in his view erased any deficiency the challenge could have had (Rallín 1991). While some of the de facto rules involved in these decisions had been ratified (mostly implicitly) by democratic authorities, they should not have been treated without any consideration as analogous to statutes. Fortunately, this use of the slogan seems to have become very exceptional. Moreover, in around 20 decisions, the challenged rule was an executive order or agency action. In a few of them, the court employed the strong Thayerian language concerning the clear and manifest incompatibility between the rule and the constitution as a prerequisite to an invalidation (e.g. Renault 2010; Zito 2014). Only a handful of those decisions involved executive orders to manage the administration, the type of executive action most clearly entitled to deference in a political system in which such action predominates.

Finally, an otherwise irrelevant bundle of decisions perhaps best evinces the problematic character of the slogan’s use. The court has repeatedly inserted it as one of the considerations to dismiss a challenge to its regulations concerning appeal briefs (e.g. Defensoría Pública 2008; Sprayette 2014; Terán 2011). Via the slogan, the court was saying that striking down its own rule concerning the length and presentation of briefs – which had been issued by the sitting justices – was as much a matter of institutional gravity as striking down a statute passed by democratically elected legislators in two houses – say, the hard-fought abortion law of 2020.

It is possible that, in some of these decisions, the court (or the Procurador General) cited the slogan unreflectively, invoking what it considered well-tested judicial doctrine concerning the presumption of rules without pausing to fully assess its soundness in the case under consideration. Availability bias may have played a role. But the fact that the slogan’s indiscriminate use was available in the minds of court justices and staff is a sign of the bastardization of constitutional law in the country – a willingness to treat similarly any rule and an uncritical disposition to resort to low-hanging doctrine – and the corrosive power of authoritarian legacies.

In a democracy still under consolidation, such as Argentina, it is normatively admissible, and to a certain extent advisable, that the court adopts a level of deference in its decision-making. However, the relevant questions seem to be deference to what and under what conditions. As noted above, a democratic legislature, most clearly in cases that do not involve the alleged violation of fundamental rights, is entitled to it, while other agencies are not so strongly or so broadly entitled to it. The court still owes both litigants and the public at least a brief explanation of its use of deferential (and non-deferential) review under democracy. More generally, while at times it may be prudent for the court to dodge certain thorny questions, as Bickel (1986) advised courts to do by embracing the ‘passive virtues’, ²¹ strict avoidance may not be the safest way for its construction of legitimacy. As Helmke and Staton (2011: 325) point out with reference to courts in Latin America, a sustained pattern of prudence may ‘risk … constructing inaccurate beliefs about judicial preferences – essentially teaching future litigants that the court is either extremely partisan or unwilling to defend rights’.

In short, the court would do well to establish a well-articulated hierarchy of rules as well as offering clear and consistent doctrine concerning the strength of its review of each of them. It would also do well to critically reflect on the language it uses and the functions it serves.

Connections

More troubling than some of the democratic court’s indiscriminate invocations of the slogan is its occasional recognition, and hence endorsement, of the last dictatorship court’s invocations. By depicting the relationship between a citing court and a cited court (Fowler et al. 2007), network analysis can be useful to analyse the extent to which the court referred to previous decisions and periods. Figure 1 displays a network of court decisions proclaiming the slogan (or an almost indistinguishable phrase) and all cross-citations in support of its use. ²² The nodes represent the court’s decisions, and they are colored according to the court’s period. The edges or links connect the nodes by showing the relationships between the decisions through the citations (Fowler et al. 2007) and are coloured on the basis of a combination of citing and cited courts. ²³ Node sizes are a function of the number of citations to them. The multi-colored nature of nodes and edges shows both that the slogan was employed across periods and copious cross-citations.

Figure 1 Supreme Court cross-citations in paragraph invoking the slogan, by court period.

Figure 2 shows instead the democratic court’s citations since 1983 to earlier cases using the slogan. This graph bundles together all the court’s democratic periods since that year. It only displays citations by the democratic court either to itself or to the last dictatorship court. In some cases, the citations were introduced by the Procurador General, who penned the opinion to which the court referred for brevity’s sake. To ease readability, citations to the periods before the dictatorship are omitted (12 per cent).

Figure 2 Democratic court’s citations (1984–2020) to democratic or dictatorship court (1976–83) in paragraph invoking the slogan.

As can be seen, references by the justices writing the majority or plurality opinion to the dictatorship court abound, with some 101 citations contained in 44 decisions mentioning the slogan under democracy to as many as fifteen decisions under the dictatorship. Remarkably, the most-cited ‘slogan’ decision under democracy was penned by the dictatorship Court (21 mentions) and involved a de facto rule. It was the case concerning restrictions on the right to strike in a country under permanent state of siege, a terrifying symbol of authoritarianism. In an indication of the court’s normalization of the abnormal after decades of irregular use, the court made half of those mentions in the context of challenges to statutes. Among the five most-cited ‘slogan’ decisions of the democratic court, three had been penned by the dictatorship court and involved the dictatorship’s regulations. While these citations have become very infrequent since the 2010s, they still feature occasionally in the court’s decision-making (e.g. Unión de Usuarios 2014; Asociación Francesa 2018). It is high time they disappeared.

VI. Conclusion

Apart from endorsing de facto rulers, the court acting under their tutelage further legitimized their legislative production by resorting to artifice, borrowing language from a type of constitutionalism that is fully reliant upon democracy. This both expressed and furthered the devaluation of statutes as the cornerstone of legislation.

While the democratic court has abandoned the most blatant manifestations of authoritarianism, more subtle connections with the previous period persist, among them its indiscriminate use of the slogan and associated canons. This partly stems from the court’s recognition of executive action in a political system strongly tilted towards this branch and continuously undergoing economic crises. By assimilating statutes and any administrative action, however, the court also had a role in that exacerbation. Moreover, the links with the authoritarian past manifested even in cases involving statutes when the democratic court cited the dictatorship court, illustrating how the depreciation of statutes – and more generally the corruption of constitutional doctrine – became normalized. It likely contributed to this result the fact that, instead of a clean break with the past, there was a level of institutional continuity at the (now democratic) court. The response to institutional breakdown may not be a continuation of old routines and practices, at least not without a careful vetting of them.

Courts and judicial review play an essential role in fragile democracies. To strengthen this role, they must operate free from undue external influence. But they also must be free from subterfuge and meticulously articulate their arguments. As countless commentators have claimed over the years, Hamilton’s dictum in Federalist #78 that all that judges have is judgment is overstated. Still, their words – including the doctrines and canons they shape – are essential, and they ought to be careful in the way they use them.