VRÜ 3 2017 50.

Jahrgang
Seite 215 – 336

Begründet von Prof. Dr. Herbert Krüger (†)

Herausgegeben von
Prof. Dr. Brun-Otto Bryde (em.), Justus-Liebig-Universität Gießen, Prof. Dr. Dr. h.c. (Univ. Athen)
Dr. h.c. (Univ. Istanbul) Thilo Marauhn, Justus-Liebig-Universität Gießen, Prof. Dr. Philipp Dann,
Humboldt-Universität zu Berlin, Prof. Dr. Jürgen Bast, Justus-Liebig-Universität Gießen, Prof. Dr.
Axel Tschentscher, Universität Bern, Dr. Karl-Andreas Hernekamp, Universität Hamburg im Insti-
tut für Internationale Angelegenheiten der Universität Hamburg durch die Hamburger
Gesellschaft für Völkerrecht und Auswärtige Politik in Verbindung mit den Regional-Instituten
des German Institute of Global and Area Studies (GIGA), Dr. Michael Riegner, Humboldt Univer-
sität zu Berlin, Dr. Michaela Hailbronner, Universität Münster, Prof. James Fowkes, Universität
Münster
Beirat: Prof. Dr. Rodolfo Arango, Bogota, Prof. Dr. Moritz Bälz, Frankfurt, Prof. Dr. Ece Göztepe,
Ankara, Prof. Heinz Klug, Madison, Prof. Dr. Kittisak Prokati, Bangkok/Fukuoka, Prof. Dr. Atsushi
Takada, Osaka.
Schriftleitung: Prof. Dr. Philipp Dann, E-mail: Philipp.Dann@rewi.hu-berlin.de

Inhalt

Abhandlungen / Articles
Stefan Salomon
Self-determination in the Case Law of the African Commission: Lessons for
Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217

Ricarda Rösch
Indigenousness and peoples’ rights in the African human rights system:
situating the Ogiek judgement of the African Court on Human and Peoples’
Rights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

Diego Werneck Arguelhes/Leandro Molhano Ribeiro

Courts as the first and only legislative chambers? The Brazilian Supreme Court
and the legalization of same-sex marriage* . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260

Bertus De Villiers
An Advisory Body for Aboriginal Peoples in Australia – one step forward and two
back? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

Berichte / Reports
Tanja Herklotz
Shayara Bano versus Union of India and Others. The Indian Supreme Court's Ban
of Triple Talaq and the Debate around Muslim Personal Law and Gender
Justice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
Buchbesprechungen / Book Reviews
Yaniv Roznai, Unconstitutional Constitutional Amendments. The Limits of
Amendment Powers (Michael Hein) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312

Mengia Hong Tschalaer, Muslim Women's Quest for Justice: Gender, Law and
Activism in India (Tanja Herklotz) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

Jacqueline Mowbray: Linguistic Justice. International Law and Language Policy

(Sabine Pittrof) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

Ben Saul / David Kinley / Jaqueline Mowbray (eds.): The International Covenant
on Economic, Social and Cultural Rights – Commentary, Cases and Materials
(Robin Ramsahye) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 323

Simon Hümmrich-Welt: Responsibility to Rebuild. Verantwortung zum

Wiederaufbau von Post-Konflikt-Staaten (Lars Müller) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326

Bibliographie / Bibliography ...................................................... 328

260

Courts as the first and only legislative chambers? The Brazilian

Supreme Court and the legalization of same-sex marriage*

By Diego Werneck Arguelhes and Leandro Molhano Ribeiro**

Abstract: Since 2012, same-sex marriage has been legal in Brazil. This, however,
was accomplished without the participation of any political branch: the only law-
making actor involved was the Supreme Court. In existing studies on the judicial-
ization of politics, lawmakers are typically the first movers, while courts function as
actual or potential veto players regarding the outcome of the legislative pro-
cess. The Brazilian Supreme Court, however, when provoked by social and political
actors under specific circumstances, has delivered decisions that: (a) established
rules in areas of law where the elected branches had not made any decisions in the
last decades; and (b) at least in how they came to be treated by the Supreme Court
itself, foreclosed further congressional or presidential lawmaking on these topics.
This paper is a case study on the conditions in which this judicial role – of
a first and only legislative chamber – has emerged in Brazil. Although Brazilian
constitutional law includes a couple of mechanisms designed for judicial review of
legislative omissions, it was the Supreme Court’s increasingly expansive reading of
its own powers – not those deliberately designed to allow for judicial legislation –
that created opportunities for social and political movements to bypass Congress
and change the constitutional status quo according to their ideals. This new role of
the Supreme Court, however, creates tension between two different justifications
for judicial intervention in politics – when judges act last, they are constrained by
not having also been the first to decide what the constitution requires and allows for
on a given issue. In same-sex marriage litigation and in several other recent cases,
however, the Brazilian Supreme Court has been trying to have it both ways.

***

* Previous versions of this paper were presented in the 2016 annual meetings of the Law & Society
Association (New Orleans, USA) and of the International Society of Public Law (ICON-S)
(Berlin, Germany). The authors would like to thank participants in both events for the helpful
debates, and especially James Fowkes, Michaela Hailbronner, Jaclyn Neo, Thomaz Pereira,
Sabrina Ragone, Scott Stephenson, and Valentina Volpe. We would also like to thank Andressa
Carvalho, Evandro Süssekind, Luna Barroso and Julia Cani for research assistance, and Victoria
Broadus for assistance in editing the manuscript.
** LL.M, J.S.D. Yale. Hauser Global Research Fellow, NYU School of Law (Fall 2017). Professor of
Law at Getulio Vargas Foundation, Rio de Janeiro, Brazil (FGV Direito Rio). Diego.werneck@fgv
.br. Ph.D. (Political Science) IUPERJ. Professor of Law at FGV Direito Rio. Leandro.ribeiro@fgv.
br.
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 261

A. Introduction

In 2011, in its Arguição de Descumprimento de Preceito Fundamental (ADPF) 132 deci-

sion, the Brazilian Supreme Court (STF) determined that the Constitution directly protects
the individual right to establish a same-sex civil union for all legal purposes. In the follow-
ing year, the National Council of Justice (CNJ) enacted a resolution – binding on all public
notaries in the country – interpreting the ADPF 132 decision as prohibiting any differentia-
tion between civil unions on the basis of the parties’ sexual orientation. As the Brazilian
Constitution states that “the law will enable the conversion of civil unions into civil mar-
riages,” the CNJ understood that, by direct implication of the STF’s ruling, same-sex part-
ners in civil unions could get married. These two decisions – one administrative, one judi-
cial – are currently the legal basis for same-sex marriages in Brazil. Such marriages have
been routine for several years now.1
This major legal transformation was accomplished without the participation of an elect-
ed branch. The only actual lawmaking actor involved was the Supreme Court. What’s more,
in technical terms, the Court did not even strike down as unconstitutional a law by which
Congress had prohibited same-sex marriages. The Brazilian Civil Code of 2002 extended
protection to civil unions formed by “the union of a man and a woman”; in doing so, it
merely repeated the exact language used in the Constitution, which said nothing specific
about same-sex marriages.
Upon enacting the Civil Code, Congress may have missed the opportunity to discuss
and adopt a constitutional amendment more in tune with the times. But, in the terms of the
Constitution as it stood at that point, the legislators were not creating anything new. Rather,
they mimicked the Constitution’s words regarding civil unions – which the text seemed to
restrict to only “a man and a woman” – and the constitutional silence on the specific issue
of same-sex marriage. At the end of the day, the Court intervened by dramatically reinter-
preting the constitutional clause, shifting from an apparent prohibition of same-sex civil
unions to a mandatory recognition of same-sex marriages.
The story of ADPF 132 presents a twist on the typical accounts of “judicialization of
politics.” The interplay between the rules enacted by legislators and the decisions handed
down by constitutional courts has been widely studied in several countries. “Judicializa-
tion” scholars have shown how in many democracies in which there is judicial review, the
lawmaking process involves strategic interactions between legislators and judges.2 As they
deliberate on a bill, foreseeing the possibility of constitutional review, legislators may

1 According to data provided by the CNJ, one year after the Council’s decision, more than 1000
same-sex couples had already gotten married. See, e.g., Regina Bandeira, Um Ano Após Norma so-
bre o Casamento Gay, Chegam a 1.000 as Uniões entre o mesmo Sexo, http://www.cnj.jus.br/noticia
s/cnj/61657-um-ano-apos-resolucao-do-casamento-gay-chega-a-1000-o-numero-de-unioes-entre-pe
ssoas-do-mesmo-sexo (last accessed on 20 August 2017).
2 For an overview of the literature on this phenomenon and the many variables it involves, see Ran
Hirschl, The judicialization of mega-politics and the rise of political courts, Annual Review of Po-
litical Science 11 (2008), pp. 93-118.
262 Verfassung und Recht in Übersee VRÜ 50 (2017)

choose to simply ignore the court as a potential threat and discuss the constitutionality of
the would-be statute on their own terms. They can, however, and often will, use the court’s
precedents to try to predict the substantive limits that future judicial rulings could impose.
Political minorities are expected to challenge these laws as soon as they are enacted. This
makes the mere existence of constitutional review mechanisms sufficient to render the
court’s preferences (as revealed by their precedents) relevant in the lawmaking process, as
legislators try to anticipate and preempt future challenges.
Judicial preferences are thus taken into account in the lawmaking process by means of
complex interactions between different actors within the political system. Typically, how-
ever, judicial action modulates – either directly or indirectly, by incentivizing strategic be-
havior – some sort of legislative action. Existing literature on the judicialization of politics
in different national and transnational scenarios includes many variations on the possible
outcomes and patterns of the interaction between “lawmakers”, broadly defined, and courts.
The legislature creates new rules; the court may disagree with them, to some extent, and
call for substantive changes to make these rules constitutional. In turn, the legislature might
anticipate these decisions in the next round of rulemaking. In these narratives, however,
lawmakers are always the first movers, even if their actions are influenced by the threat of
judicial intervention; courts, meanwhile, function as actual or potential veto players regard-
ing the outcome of the legislative process.
In the last few years, Brazilian constitutional politics has developed in ways that signal
a very different role for courts in the legislative process. The STF has demonstrated its ca-
pacity to act not simply as a veto player – a third legislative chamber – but as a first and
only legislative chamber. Social and political actors wishing to completely bypass the po-
litical decision-making process have successfully prompted the Court to deliver decisions
that: (a) established rules in areas of law where the elected branches had not taken any deci-
sions for the last decades; and (b) at least in how they came to be treated by the Supreme
Court itself, left no room at all for further congressional or presidential lawmaking on these
topics. In these circumstances, the Supreme Court completely bypassed and tentatively
foreclosed the lawmaking process. In this paper, we analyze these features of the ADPF 132
decision and explore the institutional conditions that have fostered this judicial role in
Brazil. In our reconstruction, the Court’s role as a first legislative chamber will appear as
the outcome of a set of variables, namely: the political strategies of actors outside the
Court; the institutional configuration of the Court’s powers; and the specific ways in which
the current generation of Supreme Court justices has interpreted its powers.
The rest of the article is organized as follows: In section II, we conduct a literature re-
view on the ways in which constitutional courts have traditionally been accounted for as
relevant actors in the national decision-making process. Even though there are many possi-
ble variations in how constitutional politics takes place in different countries, the common
image in these studies is the court as a third legislative chamber or as a kind of veto play-
er. In section III, we analyze the ADPF 132 decision and the role the STF played in legaliz-
ing same-sex marriage in Brazil. In section IV, we explore the conditions that made this ju-
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 263

dicial role possible. Although Brazilian constitutional law includes mechanisms allowing
for judges to act in cases of legislative omission, deliberate institutional design cannot ex-
plain the rise of the STF as a first and last legislative chamber. Rather, the specific kind of
lawsuit employed by the plaintiffs has provided the Court with unexpected opportunities to
act as a first lawmaker. These opportunities seem to have arisen from an expansive judicial
interpretation of what can be considered “legislative omission.” In section V, we conclude
by discussing how the institutional features described in section III, when combined with
certain reform-centered litigation strategies from outside actors, and when interpreted in
specific ways by STF justices, produce the conditions under which the ADPF allowed the
Court to act as a legislator.

B. Courts as legislators

Constitutional courts can play a variety of roles in the broader political decision-making
process.3 Most existing comparative constitutional scholarship focuses on these courts’
powers to review the constitutionality of statutes, administrative measures and/or interna-
tional treaties.4 Constitutional review includes two kinds of powers, each with its own polit-
ical logic: (a) the resolution of conflicts between different political decision-makers under a
vertical (federalist) or horizontal (separation of powers) relationship, with the court clarify-
ing the boundaries between these different levels of government; and (b) rights adjudica-
tion, in which “the policy-making role of courts is more apparent (…) because the logic of
seemingly neutral dispute resolution does not really mask it.”5 Most scholarship on the judi-
cialization of politics focuses precisely on this second dimension, as rights adjudication
forces courts to determine what would be the permissible content of any statute under the
constitution.6
As their legislative output is challenged before the court, congress and the president
have a variety of recourses available.7 If they decide to comply at least partially with the
court’s decisions, however, judicial participation in the lawmaking process begins to be-
come more substantive, as the court’s preferences regarding the substance of permissible

3 See, e.g., Tom Ginsburg and Zachary Elkins, Ancillary Powers of Constitutional Courts, Texas Law
Review 87 (2009).
4 According to Ginsburg and Elkins (note 4) more than 80 percent of contemporary democracies in-
clude in their sample some kind of constitutional review.
5 Ginsburg and Elkins, note 4, p. 1436: “When the court substitutes its own judgment for that of the
government or legislature, it cannot be doing anything other than policymaking.”.
6 Ginsburg and Elkins, note 4, p. 1437: “Regardless of whether the court is functioning as a bound-
ary-guarding dispute resolver or as a rights-enforcing constraint on government, a common thread
in both forms of constitutional review is judicial lawmaking.”.
7 J. Mitchell Pickerill, Constitutional Deliberation in Congress: The Impact of Judicial Review in a
Separated System, Durham 2004.
264 Verfassung und Recht in Übersee VRÜ 50 (2017)

legislation will find their way into the legislative status quo.8 In particular, when other insti-
tutional actors create a law that only partially incorporates the court’s decision, they are
putting forth a compromise solution, which will then most likely be subjected to a new
round of judicial review. To a large extent, such partial incorporations of revealed judicial
preferences take place because executive and legislative actors wish to minimize the risk of
a complete judicial rejection of their new legislative proposal. Policymaking will only re-
flect the preferences of the political branches exclusively when these branches fully ignore
the judicial threat.9
More precise empirical and theoretical accounts of this kind of judicial participation in
policymaking can be found in Stone Sweet’s work on the French Constitutional Council as
a “third legislative chamber”,10 which was then expanded to other European democracies.11
In his analyses of constitutional courts as third legislative chambers, Stone Sweet describes
constitutional politics as the outcome of close, recurrent legal-political interactions between
courts and legislatures.12 His model is built on an analysis of the actual effects of constitu-
tional review on the broader political decision-making process. Constitutional courts’ deci-
sions actually shape the content of existing policies, as judges not only “veto” laws by void-
ing them but can also change the laws’ content by establishing their own preferred interpre-
tation as binding.

8 The idea that constitutional courts can (and should) only act as “negative legislators” – i.e., they
can remove laws and provisions from the legal system but not add anything new – is credited to
Hans Kelsen’s discussion of his proposal to create a constitutional court for Austria. Whether the
existing constitutional courts can still be considered as “negative legislators” in Kelsen’s original
terms is a matter of debate (see Alec Stone Sweet, Governing with Judges: Constitutional Politics
in Europe, New York 2000, Chapter V). However, as a practical matter, as we will discuss below,
there are decisions by means of which a constitutional court effectively changes the substance of a
law by adding something that had not been foreseen by legislators (for example, when the court
reads one or more provisions into the statute in order to save it from being declared unconstitution-
al). See, generally, Alec Stone Sweet, ibid.
9 The likelihood of each of these different scenarios taking place is influenced by: (a) structural, in-
stitutional design variables, especially the design of constitutional review mechanisms (e.g., how
the court’s agenda is shaped; who can trigger abstract review; how the other branches can partici-
pate in the judicial decision-making process; who appoints the judges); and (b) contextual factors,
like the intensity of political support for the court and for constitutional democracy itself, the de-
gree of political fragmentation and the kind of policy that the political actors are trying to enact.
See Julio Ríos-Figueroa and Matthew M. Taylor, Institutional Determinants of the Judicialization
of Policy in Brazil and Mexico, Journal of Latin American Studies 38 (2006), pp. 739-766; and
Tom Ginsburg, Judicial Review in New Democracies: Constitutional Courts in Asian Cases, Cam-
bridge 2003.
10 Alec Stone, The Birth of Judicial Politics in France: The Constitutional Council in Comparative
Perspective, Oxford 1992.
11 Stone Sweet, note 8.
12 Stone Sweet, note 8 and 10; Alec Stone Sweet, Constitutional Courts and Parliamentary Democra-
cy, West European Politics 25 (2002), pp. 77-100.
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 265

Moreover, as politicians realize the risk of their policies being challenged and defeated
before the constitutional court, and because the court provides written reasons for its deci-
sions, over time, legislators will consider constitutional arguments, extracted from the
court’s previous decisions, when shaping and defending their policy proposals. This mutu-
ally reinforcing interaction between judges and politicians makes for a decision-making
process in which legal discourse is incorporated in the political process, which then be-
comes judicialized.13
The impact of judicialization on the political process can thus be direct or indirect. In
the first situation, once constitutional review is triggered, the court vetoes part or all of a
given statute. In the second, as legislators decide on the contents of a given bill, they ac-
count for the possibility of constitutional review on the basis of the court’s past decisions
and of other political actors’ future willingness to challenge the policy. This makes political
actors behave strategically, and the threat of constitutional challenges may lead to political
compromises between the legislating majorities and the political opposition.
This kind of indirect influence can take the form of “self-limitation” on the politicians’
part – when the governing coalition gives up on taking a particular legislative path due to
the threat of judicial review – or “corrective revision” when legislators rewrite a legal text
that has been reviewed by the court to respond to judicial objections and secure the bill’s
constitutionality.14 In any event, this judicialized political process is sensitive to the dynam-
ics of constitutional adjudication and the arguments it creates, as these arguments signal po-
tential judicial vetoes. The result is a process of “complex coordinate construction” of na-
tional policies.
In such a model, constitutional courts can always be conceived as a last or third legis-
lative chamber (assuming there are two chambers in the national legislature), whether their
influence on the political process is direct or indirect. They participate in the political pro-
cess after legislators have taken the initial step, and they deliberate and decide on the basis
of an actual statute adopted by a governing majority. Over time, the constitutional argu-
ments provided by the court fuel processes of self-limitation and corrective revisions of
statutes, in which the legislators seek to act in a way that will decrease the risk of judicial

13 According to Stone Sweet, note 11, p. 225, “the existence of abstract review leads (inevitably?) to
the jurisdicization of policy-making processes; as a matter of concrete policy-making impact, con-
stitutional courts behave as third legislative chambers whenever they engage in abstract review.”
Treating constitutional courts as a third legislative chamber does not necessarily entail equating
them with legislative bodies, as there are some institutional features of these courts that distinguish
their policymaking from legislative policymaking.
14 Stone Sweet, note 12,. Self-limitation can sometimes be politically beneficial to the legislating ma-
jority, in cases where it gives the government the possibility of shifting the blame for the failure to
adopt certain policies that had been demanded by the government’s constituents to the constitu-
tional courts and their legal arguments.
266 Verfassung und Recht in Übersee VRÜ 50 (2017)

action.15 In any case, the first mover is always the legislature; the court intervenes at the
end of a formal cycle that has taken place within the legislature.
This, however, is not a necessary feature of the phenomenon of judicial participation in
the political process, but rather an aspect of the specific institutional design of the countries
analyzed in existing studies – an aspect which might not be present in other legal systems.
In the next section, by looking at how same-sex marriage was legalized in Brazil, we will
discuss the possibility of high court justices directly turning their preferences into positive
law even in the absence of congressional action on a given issue – and actually because of
the lack of legislative output in that area of law.

C. ADPF 132/2011 and the legalization of same-sex marriage in Brazil

In the widely celebrated ADPF 132 case, decided in 2011, the STF constitutionalized same-
sex marriage in Brazil. The Court did so without actually striking a single provision enacted
by Congress. Instead, it interpreted the constitutional text in a way that was compatible with
only one legal treatment of same-sex unions: full equality between heterosexual and homo-
sexual couples in all spheres of life. Some of the justices acknowledged that less-than-com-
plete equality was possible – if the Constitution was interpreted as still allowing Congress
to regulate matters such as the specific procedures for a gay couple to adopt children, for
example. Still, a majority determined that the Constitution required a specific arrangement
of full equality, and therefore left no room for Congress to create any new rules in this re-
gard. Furthermore, in their reasoning, some of the justices explicitly stated that Congress
had failed to update the Civil Code as public opinion and mores had changed on these is-
sues since the Constitution was enacted in 1988.
Accusing judges of “legislating from the bench” is a familiar practice in many coun-
tries. Such accusations usually allege judicial deviation from established precedents, textual
meanings or other widely accepted methods of interpretation or binding legal materials. In
ADPF 132, however, the justices themselves presented the narrative of “judicial legisla-
tion.” In their opinions, they recognized that they were acting on a matter that, ideally,
should have been dealt with by Congress, not the Court. However, they determined that
congressional inaction had threatened fundamental rights, and therefore the Court was justi-
fied in establishing the legal regime that should be applicable to these situations.
Against the backdrop of the literature discussed in the previous sections, ADPF 132 is
an interesting case because the Court did not act as a third legislative chamber but rather as
the first and only legislative chamber, justifying its decision on the grounds of a “congres-
sional omission.” The Court presented congressional inaction as a necessary condition for
this kind of “legislative” decision on its part; yet, as we will see below, such “inaction” led
to a decision that left no further room for legislative action in this area.

15 Stone Sweet, note 9, p. 212: “Constitutional jurisprudence is nothing more or less than the lasting,
written record of a third reading by a third institution required to give its assent on a bill before
promulgation.”.
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 267

The landmark decision in ADPF 132/2011 was a major leap following a series of judi-
cial steps in the lower courts that, since the 1990s, had been granting increased protections
to same-sex couples.16 Before the 2000s, such decisions typically avoided directly address-
ing the underlying constitutional issues of equality or individual freedom. They focused in-
stead on the material implications for the right to equality of rules that prevented people of
the same gender from building a life together.17 It was not until the beginning of the last
decade that lower courts began to rule on lawsuits brought by same-sex couples explicitly
seeking legal recognition of their familial status – and even of their civil union.18
Moreover, even when such decisions were favorable to the plaintiffs, a feature of the
Brazilian legal system limited their impact on the country’s constitutional law. Brazil has a
dual-track system of constitutional review of legislation, combining both U.S.-style diffuse,
concrete review, and a set of concentrated, abstract review mechanisms located only at the
level of the Supreme Court. Formally, decisions taken by any court (including the Supreme
Court) in a concrete review context have only inter partes effect, that is, they bind only the
parties involved in the litigation. Only when the Supreme Court decides an issue by means
of abstract review mechanisms the outcome does become binding to all other courts and
public institutions. The limited scope of concrete review decisions eventually shaped the
litigation of the legalization of same-sex unions. Even if specific gay couples obtained a fa-
vorable decision from a high court, other courts could still rule differently on similar cases.
To actually change Brazilian law on same-sex unions in a stable way, therefore, it was
necessary to either institute a formal change in the legislation, or to have the Supreme Court
issue a decision within the context of an abstract review lawsuit.19 The Constitution’s word-
ing, however, seemed to block the path of ordinary legislation:

Article 226. The family, which is the foundation of society, shall enjoy special protec-
tion from the state.
Paragraph 3. For purposes of protection by the state, the stable union between a
man and a woman is recognized as a family entity, and the law shall facilitate the
conversion of such entity into marriage.

16 For an overview of the STF’s decision within the broader context of civil rights litigation in Brazil,
see Adílson José Moreira, We are Family! Legal Recognition of Same-Sex Unions in Brazil,
American Journal of Comparative Law 60 (2012).
17 Take, for example, the 1998 decision by the Superior Court of Justice (STJ) asserting that, even in
the absence of explicit legal recognition of same-sex unions, the judiciary should acknowledge the
fact that people of the same gender can join their material efforts to create a life together. See
Moreira, note 18, p. 1017.
18 Moreira, note 17, pp. 1026-1027.
19 Moreira, note 17, pp. 1032-1033 (claiming that, because “most decisions that extended full equal
protection to cohabiting couples affected only individual cases, a consequence of the limited appli-
cation of stare decisis in Brazil (...) the leaders of the gay and lesbian organizations sought other
ways to extend legal protection to all same-sex couples (…) that could substitute for legislative
measures.”.)
268 Verfassung und Recht in Übersee VRÜ 50 (2017)

This explicit wording was not accidental. As Justice Lewandowski suggested in his opinion
in ADPF 132, the drafters of the provision in 1988 were very aware that it would block the
recognition of same-sex couples’ civil unions.
Regardless of the legislative intent behind Paragraph 3, the provision’s wording indeed
presented a clear obstacle for legislation extending this protection to same-sex unions, cre-
ating an easy argument for the unconstitutionality of such protections. In this scenario, and
considering the highly decentralized system of judicial review in Brazil, creating new laws
would not be enough. Conservative judges across the country could still easily rely on Arti-
cle 226’s language to consider these new laws unconstitutional. It seemed that, as the
Supreme Court remained silent, the only path to create stable legal protections for same-sex
unions would require amending the Constitution.
On 2 September 2003, Senator Sérgio Cabral presented to the Senate the Proposta de
Emenda Constitucional (“Proposal to Amend the Constitution” or simply “PEC”) n. 70.
This PEC would have explicitly changed the wording of Article 226 of the Constitution, so
as to remove the specification that civil unions are “between a man and a woman.”20 This
formal amendment would have explicitly authorized same-sex civil unions. But Cabral’s
PEC had a brief and ultimately inconsequential life within Congress.21 In 2004, the repre-
sentative tasked with writing a report on the bill recommended that it should be debated in a
public hearing with civil society organizations [audiência pública]. Still, Senator Cabral
himself withdrew his proposal on 5 November 2006, and PEC n. 70 was sent to the con-
gressional archives.
According to the newspaper Folha de São Paulo, Senator Cabral withdrew his proposal
due to electoral strategic considerations. Folha reported that Cabral, who was running for
governor of Rio de Janeiro in 2007, backtracked in order to secure the support of Senator
Marcelo Crivella, a neo-Pentecostal evangelical preacher who had allegedly demanded the
removal of PEC n. 70 as a condition for his support.22 Cabral, who went on to win the gu-
bernatorial campaign, denied Folha’s version of the events, suggesting instead that given
how little progress his proposal had made within Congress, he had begun to consider alter-
native routes to constitutionalize the protection of same-sex civil unions.
The only other open path was judicial, and it pointed directly to the Supreme Court.
There had been a previous attempt to take the issue of the constitutionality of same-sex
marriage to the Brazilian Supreme Court, in 2007, when a group of prosecutors attempted
to convince the attorney general to initiate a lawsuit requesting the STF to rule on the issue.

20 If PEC n. 70 had been approved in its original version, Article 226, § 3º would be read as follows:
“For the purposes of protection by the state, the civil union between heterosexual or homosexual
couples is recognized as a family entity, and the law must facilitate its conversion into marriage in
the case of civil unions between a man and a woman.” (translated by the authors).
21 For more details, see Senado Federal, http://www.senado.gov.br/atividade/materia/detalhes.asp?p_
cod_mate=61093 (last accessed on 20 August 2017).
22 Folha de São Paulo, Cabral diz que não irá perder votos de gays por retirar projeto, http://www1.fo
lha.uol.com.br/fsp/brasil/fc0710200627.htm61093 (last accessed on 20 August 2017).
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 269

The attorney general chose not to take any action in this regard. However, another effort to
judicialize the issue was in the making. On 27 February 2008, former senator and then-
Governor Cabral brought his former legislative agenda back on stage, but now via a law-
suit. The State of Rio de Janeiro filed ADPF n. 132, asking the Supreme Court to interpret
Article 1723 of the Brazilian Civil Code, which repeats the wording of the Constitution:

Article 1723. The stable union between a man and a woman is recognized as a family
entity, characterized by the public, continuous and lasting sharing of one’s life to-
ward the constitution of a family.23

The plaintiff argued that, due to the mainstream interpretation of Article 1726, the State of
Rio’s homosexual civil servants could not be treated as equals with other civil servants on a
number of issues, most notably the rules concerning state-provided pensions for spouses.24
The interim attorney general filed a new lawsuit on 2 July 2009, requesting the same as
ADPF 132, with the additional request that the Court’s decision apply to all states, not only
to Rio de Janeiro.25 In both petitions, the Court was being asked to enforce the right to
equality against the wording of Article 1723, so as to assert a constitutional recognition of
the right of gay couples to have civil unions, which would then be protected by the state
like any other kind of family.
From 4 to 5 May 2011, an unanimous Court ruled that the Constitution protects stable
relationships as civil unions regardless of gender or sexual orientation.26 The justices dis-
agreed on the exact scope of this constitutional protection: three opinions explicitly stated
that it would be up to Congress to rule on how same-sex couples should be treated in a vari-
ety of situations (for example, on the procedures for the adoption of children, or on if and
how same-sex unions could be converted into same-sex marriages). Still, the Court unani-
mously agreed that Article 226 would be interpreted to include same-sex unions in its pro-
tection of family units.27

23 “É reconhecida como entidade familiar a união estável entre o homem e a mulher, configurada na
convivência pública, contínua e duradoura e estabelecida com o objetivo de constituição de
família.”.
24 Both the attorney general and the solicitor general wrote briefs in support of the plaintiff’s pos-
ition, and a large number of amicus curiae briefs were presented to the Court.
25 This suit became ADPF 178, but, upon a decision by then-Chief Justice Gilmar Mendes, it was
modified on technical grounds into ADI 4277. As opposed to the attorney general, governors, al-
though mentioned in Article 103 of the Constitution, only have “special”, restricted standing for
abstract review. In that case, this meant that the governor of Rio de Janeiro could only have stand-
ing to file such a lawsuit if he showed that the interests of the State of Rio de Janeiro were specifi-
cally affected by the denial of equality to same-sex couples. Governor Cabral dealt with this re-
quirement by pointing to the Civil Code’s implications for state rules concerning pensions for
same-sex spouses of civil servants.
26 For an extended analysis of the court’s reasoning in this decision, see Moreira, note 18, p. 1037.
27 Moreira, note 18, p. 1038: As he notes, constitutional protection of families extends to several dif-
ferent arrangements: “[C]onstitutional provisions do not distinguish between families derived from
marriage and those solemnized by the state. The social goals of the family can be achieved both by
270 Verfassung und Recht in Übersee VRÜ 50 (2017)

According to the justices, in mentioning the protection of civil unions “between a man
and a woman,” Article 226 merely gave an example of a typical family structure, which
should by no means be understood as a constitutional prohibition on the recognition of
same-sex unions.28 This is the core legal argument that seemed to bind all the justices in the
unanimous ruling: the “between a man and a woman” clause was read as an example of a
kind of family unit directly protected by the Constitution, not as an exclusion of other kinds
of families. In this way, the Court turned a potential constitutional prohibition of same-sex
unions into the constitutional requirement that these unions must be acknowledged and pro-
tected by Brazilian law.
Brazilian same-sex couples then proceeded to take two other paths to obtain full equali-
ty before the law, including the right to marry. According to Article 226 of the Constitution,
the law must recognize the possibility of converting civil unions into marriages. Therefore,
the Constitution itself grants couples in civil unions the right to convert their relationship
into a legal marriage. But was this provision also applicable to same-sex unions? Once
again, Brazil’s decentralized judicial system kicked in, with some judges accepting and
some judges rejecting this argument. Several lower courts’ decisions employed separation-
of-powers arguments that had been mentioned by the justices in the Supreme Court’s rul-
ing. They noted that, although ADPF 132 stated that same-sex unions were under constitu-
tional protection, the Supreme Court did not specify the full extent of legal implications and
possible arrangements that would be compatible with this basic requirement.29 It would
therefore be up to Congress to decide on matters such as conversion to marriage and child
adoption by same-sex couples.
The intervention of a new institution – the National Council of Justice (CNJ) – settled
the debate in the lower courts. Although not technically a court, the CNJ is part of the judi-
ciary (Article 102 of the Constitution). It was created in 2004 to oversee the administration
of the Brazilian judicial system, overseeing disciplinary proceedings against judges and the
creation of policies and rules to promote judicial efficiency, transparency and account-
ability.30 The CNJ is presided by the standing chief justice of the Supreme Court (who is
always the most senior justice who has not yet served as chief justice, elected for a term of
2 years). On 15 May 2013, Chief Justice and CNJ President Joaquim Barbosa – who, in the

same-sex and opposite-sex unions. (...) According to the Court, the institution of the family aims to
offer emotional stability to its members and an ideal set of conditions for the development of hu-
man personality. This, not reproductive capacity, makes this institution the basis of human soci-
ety.”.
28 In his opinion, the case reporter, Justice Britto’s, argued that homosexuality is a legitimate expres-
sion of the human personality that deserves legal protection, and choosing one’s partner freely is a
necessary condition for the individual pursuit of happiness. Gender and sexuality cannot be a legal
burden..
29 See Moreira, note 18, p. 1039 and for examples of such decisions, p. 1040, fn. 103 and 104.
30 See Leandro Molhano Ribeiro e Christiane Jalles de Paula, Inovação institucional e resistência
corporativa: o processo de institucionalização e legitimação do Conselho Nacional de Justiça, Re-
vista Brasileira de Políticas Públicas 6 (2016).
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 271

ADPF 132 decision, had joined Justice Britto’s majority opinion with no reservations –
asked his fellow CNJ members to vote on what would become Resolution n. 175.
Approved by 14 votes to 1, Resolution n. 175 made it binding for all public notary of-
fices and judges in Brazil to accept the possibility of converting a same-sex civil union into
a same-sex legal marriage, on the basis of Article 226 of the Constitution.31 Regarding the
Resolution’s text proposed by Chief Justice Barbosa, in its 2011 decision the Supreme
Court had already settled the matter in a way that did not require further congressional leg-
islation.32 He claimed that, as the ADPF 132 decision had clearly established that the Con-
stitution forbade Congress from treating same-sex couples differently, in all possible di-
mensions, it made no sense to wait for Congress before accepting the possibility of conver-
sion to marriage. After all, even if Congress legislated on the matter, the only legal arrange-
ment compatible with the Constitution would still be an identical set of rules for all couples.
In recent years, a handful of individual injunction decisions by Supreme Court justices
have endorsed Justice Barbosa’s and the CNJ’s stance on the meaning of the 2011 decision.
In 2015, for example, Justice Carmen Lúcia, in an individual decision on an appeal, ruled
that a lower court judge accept the adoption of a child by a same-sex couple under the same
rules applicable to heterosexual couples.33 While the CNJ resolution itself has been chal-
lenged before the Supreme Court by means of an abstract review lawsuit, the existing indi-
vidual rulings confirming the expansive reading of ADPF 132 give very little indication, at
this point, that a majority of justices would be willing to revisit and restrict the scope of that
decision.34 At the time of writing, more than six years after same-sex marriage was made

31 Conselho Nacional de Justiça, Resolução que disciplina a atuação dos cartórios no casamento gay
entra em vigor amanhã, http://www.cnj.jus.br/noticias/cnj/24686-resolucao-que-disciplina-a-atuac
ao-dos-cartorios-no-casamento-gay-entra-em-vigor-amanha (last accessed on 20 August 2017).
32 “CONSIDERANDO que o Supremo Tribunal Federal, nos acórdãos prolatados em julgamento da
ADPF 132/RJ e da ADI 4277/DF, reconheceu a inconstitucionalidade de distinção de tratamento
legal às uniões estáveis constituídas por pessoas de mesmo sexo; CONSIDERANDO que as referi-
das decisões foram proferidas com eficácia vinculante à administração pública e aos demais órgãos
do Poder Judiciário; CONSIDERANDO que o Superior Tribunal de Justiça, em julgamento do
RESP 1.183.378/RS, decidiu inexistir óbices legais à celebração de casamento entre pessoas de
mesmo sexo; CONSIDERANDO a competência do Conselho Nacional de Justiça, prevista no art.
103-B, da Constituição Federal de 1988; RESOLVE: Art. 1º É vedada às autoridades competentes
a recusa de habilitação, celebração de casamento civil ou de conversão de união estável em casa-
mento entre pessoas de mesmo sexo. Art. 2º A recusa prevista no artigo 1º implicará a imediata
comunicação ao respectivo juiz corregedor para as providências cabíveis. Art. 3º Esta resolução
entra em vigor na data de sua publicação.”.
33 Portal Brasil, STF reconhece adoção de criança por casal homoafetivo, http://www.brasil.gov.br/ci
dadania-e-justica/2015/03/ministra-do-stf-reconhece-adocao-de-crianca-por-casal-homoafetivo
(last accessed on 20 August 2017).
34 ADI 4966, filed by the Christian Social Party (Partido Social Cristão, PSC) in June 2013. The
Court has yet to schedule this case for judgment. Given how easily the Court and even its individu-
al justices can prevent a decision on any case considered inconvenient or not a priority, however,
there is no guarantee it will go back to this case anytime soon. For a discussion of timing control
272 Verfassung und Recht in Übersee VRÜ 50 (2017)

constitutionally required, the legislative status quo shaped by the Court and the CNJ has re-
mained undisturbed.

D. The first, the last, and everything: a new role for the Supreme Court?

Under the discursive mantle of reviewing “legislative omissions,” in ADPF 132, the STF
actually functioned as the first and (even if tentatively) last legislator, by implicitly reading
into the Constitution a requirement that all kinds of unions between people of any gender or
sexual orientation should be regulated under exactly the same set of rules. Some of the jus-
tices explicitly phrased the issue in terms of legislative “inaction,” and several of the opin-
ions even cite bills and reform proposals on same-sex marriage that had stopped in their
tracks in Congress.35 However, as we noted, there is an explicit constitutional provision us-
ing the words “man” and “woman” in conjunction with the protection of stable unions,
making it harder to see how Congress could have possibly legislated same-sex unions into
the Brazilian legal system without actually amending the Constitution. Accepting the dis-
course of “legislative omission” in this case would force us to consider that a failure to
amend the Constitution can be treated as an unconstitutional omission – that is, that one
part of the original constitutional text can be used as a standard against which another origi-
nal constitutional provision could be considered so problematic as to require amending as a
constitutional (not just political) matter. This then leaves us with the question of if and how
we can treat different parts of the same, original constitutional text as incompatible with
one another.
In spite of the talk of “legislative omissions,” however, the Court’s decision presents
Congress with only one possible arrangement – full and complete equality – that would be
compatible with the constitutional principle of equality. In this picture, even if Congress
had created a legal framework to protect same-sex civil unions and even same-sex mar-
riages, any such arrangement that established any difference in the rules applicable to same-
sex and heterosexual couples would be unconstitutional. Several of the justices – including
the reporter’s opinion – go as far as framing this understanding of equality as connected to
the unamendable provisions of the Constitution, signaling to Congress that future amend-

mechanisms in the STF, see Diego Werneck Arguelhes and Ivar A. Hartmann, Timing Control
without Docket Control: How Individual Justices Shape the Brazilian Supreme Court’s Agen-
da, Journal of Law and Courts 5 (2017), pp. 105-140.
35 See, for instance, Justice Marco Aurélio’s opinion: “Here is the fundamental question in this de-
bate: if the public and lasting sharing of a life together, with the goal of forming a family, between
people of the same sex should be admitted to be a family unit under the Law of the Land [Lei
Maior], considering the legislative omission (...) Would it be possible to include, under this legal
category, a situation that was not explicitly dealt with by the legislature (...)? Would this be going
beyond the limits of the judicial power? The answer to this last question is (...) unequivocally neg-
ative.” For other references in the opinions to the reform proposals that had been “stuck” in
Congress, see Rachel Nigro, A decisão do STF sobre a união homoafetiva: uma versão pragmática
da linguagem constitucional, Direito, Estado e Sociedade 41 (2013), p. 166.
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 273

ments attempting to remove or restrict the protection of same-sex couples could themselves
be considered unconstitutional.
Regardless of one’s substantive view of the principle of equality, there would be other,
less ambitious roles for the Supreme Court to play in such a scenario so as to increase legal
protection for same-sex couples.36 The Court could have emphasized its role as a first
mover by signaling to Congress that a constitutional amendment was needed, or that it was
prepared to read Article 226 in a more flexible light so as to uphold legislative measures to
protect same-sex unions and marriages.37 It could have even created a set of rules – includ-
ing full equality – that could be tentatively changed by Congress. These adjustments would
themselves be challenged before the Court in future cases, and the Court could then gradu-
ally influence public policy in this area by exercising a veto role and allowing its jurispru-
dence to shape new rounds of legislation. Judicial intervention as a first move is typically
justified because it is not the last one; conversely, and in contrast, judicial intervention as a
last word is typically justified on the basis that the court is deciding on something that
Congress actually did. In ADPF 132, however, the Supreme Court had it both ways.
This is a very ambitious role for an apex court to play in the political process, even in
regard to fundamental rights. But what set of conditions allowed the Court to act as a first
and last legislative chamber in this fashion? Three potential elements can be extracted from
our analysis of ADPF 132. The first is litigation itself, that is, the availability of cases that
presents this opportunity to the Court, in the form of ADPFs brought before the justices in
which they are asked to “fix” a constitutional problem that has been ignored by Congress.
Litigation is shaped by social and political actors’ expected benefits (including but not limi-
ted to an actual victory) to be obtained from both the legislative and the judicial arena.38
These incentives for judicalization are shaped by a second variable – institutional design –
which, in turn, is shaped by yet another variable: the Court’s interpretations of its own pow-
ers.39
Consider how these three factors interacted in leading to the decision in ADPF 132: As
we discussed in section C above, social movements and political actors moved to the ab-
stract review arena after realizing that the political decision-making process was closed to
them. They would arguably need a constitutional amendment to change the legal status quo

36 See, e.g., the comparative discussion in Sabrina Ragone and Valentina Volpe, An Emerging Right
to a 'Gay' Family Life? The Case Oliari v. Italy in a Comparative Perspective, German Law Jour-
nal 17 (2016).
37 See, e.g. Michaela Hailbronner, Overcoming obstacles to North-South dialogue: Transformative
constitutionalism and the fight against poverty and institutional failure, Verfassung und Recht in
Übersee 49 (2016), pp. 253-262: the “engagement remedy” suggested by Hailbronner in cases of
“institutional failure” as a broader category of unconstitutional omission by the state.
38 See Julio Ríos-Figueroa and Patricio Navia, The constitutional adjudication mosaic of Latin
America, Comparative Political Studies 38 (2005), pp. 189-217; Matthew M. Taylor, Judging poli-
cy: courts and policy reform in democratic Brazil, Stanford 2008.
39 Diego Werneck Arguelhes, Poder não é querer: preferências restritivas e redesenho institucional no
Supremo Tribunal Federal pós-democratização, Universitas Jus 25 (2014).
274 Verfassung und Recht in Übersee VRÜ 50 (2017)

according to their preferences and allow for the protection of same-sex unions. But what
instruments did the legal system give them to trigger judicial review in such a scenario of
legislative silence, rather than legislative action? This had been a topic of explicit debates in
the constitutional convention of 1987-88, and the Brazilian Constitution contains a set of
mechanisms that explicitly empower the Supreme Court to exercise constitutional review of
congressional inaction. These mechanisms – the Mandado de Injunção (MI) and the Ação
Direta de Inconstitucionalidade por Omissão (ADO) – were new institutional solutions to
the old problem of potentially transformative rights provisions that never made any practi-
cal difference, due to being considered “non-self-applicable” – and therefore not judicially
enforceable – until Congress and the president approved enabling legislation.40
As an abstract review lawsuit, the ADO may be filed by a selected few political actors
and institutions; an MI, in contrast, may be filed by any plaintiff in a situation where legis-
lative or presidential omission has “disabled” the exercise of a fundamental right. Accord-
ing to the STF’s interpretation of these mechanisms, however, they are of limited use to liti-
gants expecting the Court to act as first legislative chamber.41 The ADO may only be used
to ask the Court to certify that there is an unconstitutional omission and to issue a corre-
sponding warning to Congress. The MI, in contrast, has been actually interpreted in recent
years as allowing the STF to create provisional rules to neutralize the omission while
Congress does not legislate. However, the MI can only be filed against an “omission” in a
narrower sense, as defined by an explicit constitutional clause – for example, when a given
constitutional guarantee explicitly mentions the need for Congress to make enabling laws
for its full application, and no such a law has been passed.42
Although in establishing rules for the enforcement of the special retirement conditions
for some categories of public servants the Court does act as a legislator, this area of law
provides rather limited opportunities for the Court to engage with constitutional politics as
described by Stone Sweet and others. Moreover, the MI is very circumscribed by the consti-
tutional text itself: it can only be filed when the Constitution explicitly states in a provision
that some kind of specific law of Congress or another lawmaking authority is required be-
fore a fundamental right can be exercised. That is, the Constitution itself creates very spe-
cific areas in which the failure to enact a specific statute could lead to judicial review.

40 See Alexandre Araujo Costa and Henrique Augusto Figueiredo Fulgêncio, O mandado de injunção
na Assembleia Nacional Constituinte de 1987-1988, Revista Estudos Institucionais 2 (2017), pp.
818-870; Luís Roberto Barroso, O direito constitucional e a efetividade de suas normas: limites e
possibilidades da Constituição brasileira, Renovar 2009.
41 For a detailed discussion of how the Supreme Court of the 1990s read the Constitution with re-
strictive lenses when it came to establishing the boundaries of constitutional review, see Arguelhes,
note 40; Diego Werneck Arguelhes, Old Courts, New Beginnings: Judicial Continuity and Consti-
tutional Transformation in Argentina and Brazil, New Haven 2014.
42 The Court had been tweaking its understanding of the MI for a decade before finally setting it
aside in 2007. For an account of the Court’s transformation of its jurisprudence in the MIs 708 and
712, see Alice Voronoff, Ativismo judicial e democracia: por uma teoria eclética do mandado de
injunção, Revista Brasileira de Direito Público 10 (2012).
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 275

Institutional design has been relevant in Brazil in making this new judicial role possi-
ble, but only insofar as it has been connected to the Court’s expansive readings of its own
powers. Beyond the MI and the ADO, another mechanism in Brazilian constitutional law –
the one that has been central to our analysis here – has unexpectedly allowed the Court to
engage in policy areas before Congress or the president have enacted any rules: the Ar-
guição de Descumprimento de Preceito Fundamental (ADPF). Although the ADPF is men-
tioned in the original constitutional text enacted in 1988, it was not until 1999 that this type
of lawsuit was given a more concrete form and made available for litigation.43 Article 103,
§ 1o simply states that “a claim of non-compliance with a fundamental precept deriving
from this Constitution shall be examined by the Supreme Federal Court, under the terms of
the law.” In 1998, Congress enacted Law n. 9882, designing the ADPF as an abstract re-
view lawsuit, with the same standing rules as the other abstract review lawsuits before the
Supreme Court.44
In contrast to the generic abstract review lawsuit created by the Constitution (“Ação Di-
reta de Inconstitucionalidade,” or ADI), however, the ADPF can be used to challenge any
“acts of the public powers” (a broader array of possible targets than just federal or state
statutes) that violate “fundamental precepts” of the Constitution (a narrower substantive
standard of review). Although the ADPF was not directly created as a mechanism to deal
with legislative omissions in a broader scope than the MI and the ADO, it has been used in
some recent, high-profile cases by the Supreme Court to do exactly that. Law n. 9882/99
says that the ADPF can be filed against any “acts of the public powers” but the Court has
read it as being employable against inactions and omissions as well.45
Judicial interpretation of the Court’s own powers becomes even more decisive precisely
because the scope and uses of the ADPF have been so disconnected from textual limita-
tions. In contrast to the more circumscribed omission in the case of the MI, which can only
be filed when a constitutional provision explicitly requires the creation of a specific law, the
“omission” here is in the eye of the (judicial) beholder. The Supreme Court justices have
employed the language of legislative omission in cases that could perhaps be just as reason-
ably described as cases of legislative decision. For example, the new Civil Code, enacted as
recently as 2002, did not change existing policies regarding same-sex marriage. The jus-
tices described this as an omission, but those legislators would arguably see themselves as

43 See Dimitri Dimoulis and Soraya Lunardi, Curso de processo constitucional: controle de constitu-
cionalidade e remédios constitucionais, 2013, p.160: observing that the ADPF was “inactive” until
1999, and that even after that it had “limited relevance”.
44 Article 103 of the Constitution. It was the STF’s own interpretation of the new constitutional text
in the early 90s, however, that established that existing abstract review mechanisms could not be
used against pre-constitutional statutes. See Arguelhes, notes 40 and 42.
45 Dimoulis and Lunardi, note 44, p. 166; Gilmar Ferreira Mendes, Controle de constitucionalidade:
uma análise das leis 9868/99 e 9882/99, Revista Diálogo Jurídico 11 (2002), pp. 76-78. Dimoulis
and Lunardi, note 44, p. 167: They also note how the scope of the ADPF is expanding in other
directions as well – for example, the STF has established that, in some circumstances, it is possible
to challenge judicial decisions by means of an ADPF.
276 Verfassung und Recht in Übersee VRÜ 50 (2017)

presenting the nation with a decision not to change the status quo: the existing laws are fine
as they are, and we, legislators, do not currently think we need to accommodate same-sex
partnerships within the framework of marriage and civil unions.
The Court’s majority decision in ADPF 132 will most likely be the subject of new
rounds of legislation, perhaps even of new judicial decisions, even though the justices pre-
sented their decision as actually precluding any further lawmaking by Congress. It is still
early to determine whether the Court will be successful in moving from first legislator to
first and last legislator on the issue of same-sex marriage, and it is very likely that Congress
will react. For the time being, however, ADPF 132 (as interpreted by the CNJ) stands as the
law of the land: as several legislative proposals to legalize same-sex marriage came to a halt
within Congress, a judicial decision and an administrative resolution made this change pos-
sible in Brazil. It would be surprising if activists connected to other social and political
causes missed this event, and the kind of opportunity for strategic litigation that it exempli-
fies.

E. The rise of an “all-purpose” constitutional review procedure?

The literature on the judicialization of politics across the globe has mostly been developed
by observing the impacts that constitutional courts had in the policy-making process when
deciding on rules that had just been created by the political branches. Some studies do focus
on constitutional review of legislative omissions more generally but these scenarios usually
consist of the court forcing the political branches to comply with existing rules.46 Existing
scholarship on judicialization in Brazil also focuses on cases in which the STF had shaped
policies that had just been adopted by Congress or, to a much lesser extent, the president.47
There is a reason why these cases are the centerpiece of judicialization studies: when
the constitutional court enters the policy-making process after a new statute has been enact-
ed, it is at its most publicly and politically exposed point. When it acts as a third legislative
chamber, the court is deciding on an issue that is still very much on the governing majori-
ty’s agenda. The image of the court as a third legislative chamber, shaping law directly (by
vetoing statutes or reading new clauses into them) or indirectly (as legislators start to be-
have strategically before the threat of judicial review), is still central to understanding judi-
cial power in the political process. However, in recent years, some of the most high-profile
cases decided by the Brazilian Supreme Court have taken a different form: the Court self-
consciously stepped in to enact rules in areas in which Congress had not legislated but ac-
cording to the justices should have.
In this paper, we have discussed ADPF 132 as an example of this phenomenon – but it
is not the only one. Other comparable decisions have been taken in the last few years. In

46 See, in the case of Colombia, Rodrigo Uprimny, Judicialization of politics in Colombia: cases,
merits and risks, Sur - Revista Internacional de Derechos Humanos 3 (2007), pp. 53-69.
47 See, e.g., Taylor, note 39.
Arguelhes/Ribeiro, Courts as the first and only legislative chambers? 277

2012, for example, the STF decided, also by means of an ADPF, that the abortion provi-
sions of the 1940 penal code could only be reconciled with the Constitution by creating an
exception for cases in which the fetus has no brain.48 While a few justices also pointed to
legislative omission in that case, remarking that Congress had failed to “adjust” the old pe-
nal code to recent technological changes in prenatal diagnostics, the Court’s decision estab-
lished that forcing a woman to take such a pregnancy to term would amount to a violation
of human dignity. Once again, behind the language of legislative omission we find the jus-
tices determining exactly what the legislation should look like on that specific issue in order
to be constitutional – without any previous legislative controversy created by any measures
taken by Congress.
A more recent example is ADPF 378/2015, the landmark decision made by the STF at
the beginning of President Dilma Rousseff’s impeachment procedure in the House of Rep-
resentatives. A left-leaning political party (Partido Comunista do Brasil, PCdoB) filed a
lawsuit against Federal Law n. 1079/50, which regulated impeachment procedures and de-
tailed impeachable offences within the federal government. PCdoB argued that Law n.
1079/50 was not fully compatible with the Constitution, as it gave Congress too much lee-
way to define an impeachable offense, and also established trial procedures that were insuf-
ficiently protective of due process of law and the separation of powers. Law n. 1079/50,
however, had been applied in the early 1990s in President Fernando Collor’s impeachment
trial, when its basic compatibility with the Constitution had already been asserted by both
Congress and the Court. Just as with ADPF 132, it made little sense to frame the issue as a
congressional omission. What the plaintiffs wanted (and got, to a large extent) from the
Court was to rewrite the impeachment procedures according to the justices’ view of what
the Constitution requires.49
More recently, the Court started to decide ADPF 402, filed by the political party Rede
Sustentabilidade (REDE) in May 2016. REDE argued that because the Constitution deter-
mined that a president should be suspended from office for 180 days when being tried for
criminal offences, it should also be interpreted as prohibiting any politician who is a defen-
dant in a criminal lawsuit before the Supreme Court from heading one of the branches of
government. That is, no one facing criminal charges before the STF could be the speaker of
the house, the president of the senate, or the chief justice of the Supreme Court, and anyone
who became a defendant this way would have to be suspended from office as soon as the
STF accepted to hear such charges.50 Before REDE filed this lawsuit, such an argument
was unprecedented in Brazilian politics or constitutional law. There was no hint of public or

48 ADPF 54.
49 ADPF 378. See Juliano Zaiden Benvindo, Institutions Matter: The Brazilian Supreme Court’s De-
cision on Impeachment, http://www.iconnectblog.com/2015/12/institutions-matter-the-brazilian-su
preme-courts-decision-on-impeachment/ (last accessed on 15 October 2017).
50 Estado de São Paulo, Réus, Presidentes, Senadores e Ministros, http://politica.estadao.com.br/blog
s/supremo-em-pauta/reus-presidentes-senadores-e-ministros-cronica-da-adpf-402 (last accessed on
20 August 2017).
278 Verfassung und Recht in Übersee VRÜ 50 (2017)

scholarly debate or mobilization in that direction. Although the Court has not reached a de-
cision on ADPF 402 yet (one of the justices interrupted the trial asking for more time to
consider the issue), the opinions announced so far suggest a majority in support of the
plaintiff’s argument.51
These examples point to a transformation in the dynamics of the judicialization of polit-
ics in Brazil.52 From an extension of the political arena (a “third chamber” or a “veto play-
er”), the Supreme Court has been moving to the first stage of policy-making debates. This
new role includes being the first mover in debates on how the Constitution should be
changed to solve problems that have just appeared in the national public sphere, on which
Congress had neither legislated or actively considered the issue and decided to leave the
legislative status quo untouched. Congressional action was made unnecessary in ADPF
132/2011, but that was still the outcome of social and political forces that, for years, had
organized and pressured legislators to amend the Constitution. To some extent, the same is
arguably applicable to ADPF 54, when the Court decided in favor of measures supported by
reproductive rights and public health activists in cases before the lower courts. In the more
recent examples we discussed above, however, there was no previous stage of attempting to
change the law through Congress before taking the "omission" to the Court.
This suggests that the backing of a social movement or the existence of organized, sus-
tained political action might not be a necessary condition to make the STF act as a first and
last legislative chamber. It would be surprising if activists connected to other social and po-
litical causes had not yet realized the kind of opportunity for strategic litigation that the
ADPF provides. We believe that the future of judicial politics in Brazil will be shaped by
how litigants and the Court itself deploy the powerful tool of the ADPF as a mechanism to
identify a problem Congress has not yet decided – and tentatively settle it once and for all,
with judges as the first and only legislators.

51 In the Brazilian Supreme Court, deliberations can be suspended by a series of informal mechan-
isms even after a majority of the Justices have already announced their opinions. The official justi-
fication is that, until the final result is formally announced, any Justice can in principle change his
position or adjust his opinion to account for a new argument. For a discussion of strategic uses of
such mechanisms to control the timing of the STF’s decisions, see Arguelhes and Hartmann, note
35.
52 For a development of this argument in dialogue with the literature on judicial politics in Brazil, see
Diego Werneck Arguelhes and Leandro Molhano Ribeiro, Criatura e/ou Criador: transformações do
Supremo Tribunal Federal sob a Constituição de 1988, Revista Direito GV 12 (2016), pp. 405-440.
.