2019 Wright Lecture – Mark Tushnet – “Institutions for Protecting Constitutional Democracy”
Articles,  Blog

2019 Wright Lecture – Mark Tushnet – “Institutions for Protecting Constitutional Democracy”

(audience applauds) Thank You, Dean Iacobucci
and Professor Su. Thanks to the Dean for inviting
me to give this lecture, and to the faculty as a
whole for its hospitality during my two weeks here
teaching an intensive course. I will say a little more detail about the nature of this
project in a moment. But I just wanna launch
into the discussion directly with the observation that if
I don’t blather on too long, I’m happy to have questions and discussion after I’ve concluded my main presentation. Particularly because
this is an early stage of what I hope to be a
more extended project. So let me frame the project. Constitutional theory,
dating from at least the 18th century, identified
three functions of government, each associated with
what we would now call a distinctive institution. Making rules to regulate
and guide behavior, done by the legislature. Implementing those rules,
done by the executive. And resolving disagreements
about the meaning and applicability of the rules, which is done by the judiciary. In this Montesquieuian framework, each branch had its own function,
and was to be as insulated from the other branches as possible. I think the favorite
quotation in support of this is from the Massachusetts
Constitution of, I think, 1780. In the government of this Commonwealth, a legislative department
shall never exercise the executive and judicial
powers or either of them. The executor shall never
exercise the legislative and judicial powers, or either of them. The judicial should never
exercise the legislative, executive powers, or either of them, to the end that it may
be a government that laws and not of men. Now, in the late 19th
and early 20th century, this hermetic separation of
powers came under pressure with a development of modern
administrative bureaucracies. Those bureaucracies blended
lawmaking, law application, and law interpretation. And constitutional theory adapted
by treating administrative bureaucracies as a fourth
branch of government. I think the label probably
was initially attached in the United States by James Landis. Now we should note two
things about this new branch. First, it didn’t actually
perform any functions not previously performed by
the Montesquieuian branches. It was law developing, law
applying, and law interpreting, but nothing new. And second, and perhaps related, if you look at constitutions
around the world, they actually rarely
identify the administrative as a distinctive branch of government on a par with the Montesquieuian branches. I edited a handbook in which
the author of the section on the administrative branch did a survey. And there just aren’t that
many express recognitions of the administrative
as a separate branch. Now for present purposes, the next development in
constitutional theory, the constitutional theory
of government powers, came in the 1920s with Hans
Kelsen’s identification of a previously overlooked function, and his conceptualization
of a new institution to perform that function. The function was the
protection of the Constitution. And the institution was
a Constitutional Court separate from the ordinary judiciary. Now Kelsen’s innovation spread widely over the course of the 20th century. But by the late 20th century, Constitution designers came to understand that the function of
protecting the Constitution, or at least the function of protecting constitutional democracy, could be performed by other institutions, some of which were better placed than the Constitutional Court, to protect specific conceptions
of constitutional democracy. And I’ll say a bit about
some of those institutions in a moment. The constitution of South Africa provides a convenient
formulation in chapter nine, which I’ve adopted for my title. Chapter nine identifies these institutions as institutions protecting
constitutional democracy. In my shorthand, I call them IPDs, because IPD is easier to say than IPCDs. So I’ll just be calling them IPDs. Now, this lecture is the
second in a series of essays in which I explore aspects of
this new branch of government. The first essay used a case study of anti-corruption
investigations in South Africa, and the constitutional analysis associated with those investigations
as a vehicle for opening up some of the themes examined today. Which therefore should be
seen as offering something like a more extended
conceptual and historical foundation for the larger project. I’m gonna start in the
next section of the talk with an examination of the reason IPDs perform a distinctive
function, and then turn to constitutional and
statutory antecedents to them. After that, I’ll return to
the Constitutional Court as an institution to discuss why supplementing a Constitutional
Court with other IPDs might be desirable, as a matter
of constitutional design. And I’ll also talk about the relationship between the Constitutional
Court and the other IPDs. Before concluding, I’ll
discuss whether it’s useful to think of these institutions as a new branch of government, rather than as a sort of
conjuries of institutions with no more than a family resemblance. And then I’ll wrap up by asking
whether, or maybe better, how much constitutional
design through the use of these institutions can
actually do to preserve democracy. Okay, so what is distinctive
about these institutions? I’ve described the Kelsenian
Constitutional Court as an institution designed
to protect democracy. I’m now going to formulate my statements in terms of Kelsen’s thinking. I have come to understand
that people who are scholars of Kelsen’s work have questions about my characterization of his thinking. And that I attributed to him
actually doesn’t matter much. But I just want to forestall the objection that Kelsen actually didn’t
think what I’m about to say. I’m offering an account
of what might be said about why the the Constitutional Court was a distinctive institution. Now, with that, I will
start using Kelsen argued kinds of language. Believing that the British system of political constitutionalism
was inadequate, either in general or under the conditions of post World War I Europe, Kelsen argued that a
Constitutional Court was needed to ensure that democracy did not crumble, when conflicts between Parliament
and the executive arose. For Kelsen, it was obvious that in a world with political parties, the
legislature could not be trusted to respect the boundaries
between the domain of legislation and the domain of execution. And in a world of political parties, that political contention could not supply the necessary constraint either. And it was equally obvious to him that the same problems
affected the executive, with respect to the
boundaries of its power. The Kelsenian Constitutional
Court then was a response to the perceived problem,
as I characterize it, of conflicts of interest that
might undermine democracy. Conflicts of interest
between institutions. That once the problem
of conflict of interest is brought to the surface,
we can see such conflicts well beyond the legislative
executive conflict identified. The most obvious one comes with the rise of a human rights culture. There there’s a conflict between
the government as a whole, and rights holders. And Kelsenian Constitutional Courts, which were initially designed
simply as border policing institutions between the
executive and the legislature, came to absorb the function
of protecting human rights. These, though, there are many
other conflicts of interest that threaten democratic stability. The most obvious perhaps involves the law of politics itself. The qualifications of
parties for the ballot. Determining whether an
election was conducted reasonably fairly and freely. Drawing district lines and the like. And electoral commissions and courts, and boundary commissions
are, for this reason, fairly described as institutions
protecting democracy. Anti-corruption agencies
are another candidate. They protect democracy. Or this is not definitional
in connection with democracy, I think, but sort of an empirical truth that people have come to appreciate. Anti-corruption agencies protect democracy because pervasive corruption
undermines public confidence that the government the people elect, actually is worth defending. And we can’t be confident that, again, in a world of political parties, legislatures or executives
will investigate high-level corruption adequately. When both the legislature
and the executive are controlled by the same
party, everybody has an interest in sweeping corruption under the rug. And when they’re controlled
by different parties, the anti-corruption investigation may well be infected by partisanship. An executive will use
such an investigation to take out its political enemies, on either trumped-up
charges of corruption, or more troublesome, by using aggressive, but not unreasonable
definitions of corruption to go after actions by
political opponents, that the executive leaves untouched when committed by its
political supporters. And, of course, the same is true of legislative investigations
of executive corruption. Now, I have to say a bit about
the best design for IPDs. It does have to respond to
the problem of conflicts of interest, but it also
has to take into account the role these institutions play in the ongoing functioning of
constitutional democracies. The general solution, modeled
on designs for selecting judges who engage in
enforcing the Constitution, is to use a combination of
independence and accountability, though the balance may vary
from one IPD to another. And the design question is important because conflicts of
interest take many forms. Here I wanna distinguish
between two general categories. The first is intratemporal
or contemporaneous conflicts. And second, is intertemporal ones. The examples I’ve used so
far are mostly examples of contemporaneous conflicts of interest. Today’s legislature has an interest in engaging in overly
aggressive investigations of executive corruption, for example. But democracy can be
threatened by intertemporal conflicts of interest as
well, by decisions taken today that threaten democracy in the future. And the two examples I use
are, I think, quite obvious, and are sometimes recognized
by constitution designers. The first is environmental protection. Ensuring that democracy will not collapse under the pressure of
emergencies created by a failure to address and plan for
environmental change, pits today’s voters against the future. We have to bear costs today that will benefit only
people in the future. The second example involves central banks, whose general mission is to
ensure that today’s politicians don’t adopt fiscally
irresponsible policies today, to garner public support
and achieve reelection, where those policies
will lead to politically destabilizing inflation in the future. That’s a standard account of why we have independent central banks. Now these intertemporal
conflicts are real. The problem is that we
might be able to identify so many of them, and maybe even so many contemporaneous conflicts of interest, that IPDs come to dominate
ordinary policymaking. That is, if we define
conflicts of interest broadly, the less than fully accountable IPDs will absorb an extremely
large swath of public policy. Now, my current thinking
about this difficulty, which I confess I’m quite
open to reconsidering, is that good constitutional
design would address it by identifying something like thresholds for the institution’s jurisdiction. The thresholds would, of course, differ depending on the nature
of the conflict of interest. For anti-corruption, we
might limit the agency’s jurisdiction to high-level corruption, hoping that political parties will find it in their own interest to
eliminate street-level corruption. And there are theoretical accounts according to which authoritarian parties, the Chinese Communist
Party is a classic example, do have an interest in rooting
out street-level corruption, while tolerating high-level corruption. Although even in such systems, the dominant or authoritarian party can use investigations of
street-level corruption to weaken their political enemies, and strengthen their political friends. That appears to be the
experience in China, for example. For environmental protection,
we might limit the agency’s jurisdiction to matters closely
connected to climate change, excluding protection of
endangered species, for example. Now, whether the idea of
jurisdictional thresholds can be developed in enough
detail to prevent IPDs from taking over too
much of the policy space, is for me an open question. I wanna turn now to the antecedents, to the modern IPDs, with the aim of identifying some disadvantages with some of them. Constitutions always draw
upon pre-existing forms, even as they transform them. You can find many historical
antecedents to IPDs. Sun Yat-sen proposed
adding a Control Yuan, roughly an auditing bureau
and an Examination Yuan, roughly a civil service system, to the Montesquieuian system. And he theorized this
expressly as an alternative to the Montesquieuian system. He thought these two things
could not be well performed by the Montesquieuian branches. Other forms of audit
courts and auditing bureaus date back again at least
to the 17th century. And more recently, the Nordic
nations designed Ombuds offices for citizen provoked oversight of how the laws were being administered. And those offices, too, can sit outside the Montesquieuian system. Now, I’m gonna mention something about Sun Yat-sen’s account, which I draw from a
Chinese scholar writing many years ago now. It’s the best I could find
in the time available. According to this scholar,
Sun Yat-sen was convinced that if the censorial power, which is how this author
characterizes the Control Yuan, were in the hands of the legislature, it would be abused when
the parliamentary majority was not in power in government, and purposely neglected
when the Parliamentary power majority was in power in government. The conflict of interest
that I’ve identified. Sun Yat-sen’s theorization of his scheme, was compatible with, or maybe a precursor of, my analysis. As far as I know, though, the creation of these
various institutions, audit courts and the like, did not generate beyond this
observation by Sun Yat-sen. Didn’t generate wider reflection on their relation to preserving democracy. They were treated simply as examples of good government innovations. Again, I think I need
to look into the sort of early literature on Ombuds
offices to confirm that. But that’s my general impression. I say early literature because as the idea of chapter nine institutions
sort of comes into visibility, you might re-theorize the Ombuds offices in these kinds of conflict of interest terms. Okay, so now I’ve been talking about some constitutional antecedents
to these institutions, statutory antecedents to them also abound. And examining some of them brings out some important questions of design. As a first cut, I tell my students that I think in binary terms. So as a first cut, we can
distinguish between ad hoc and permanent statutory institutions that perform the functions that
constitutionalized IPDs do. Examples of ad hoc institutions
are special councils to investigate some defined set of events, as in the United States
Commissions of Inquiry into various events that have taken place. Canada and Great Britain
use Commissions of Inquiry. And a case that I find just
interesting on the facts, the commission created
in the United States to resolve disputes over votes in the 1876 presidential election. I’ll describe that in more
detail in just a moment. These institutions are
created after the event, the scandal, the police
misconduct, the election itself. And that fact alone creates
a problem of design. Permanent institutions are created under a veil of ignorance,
because no one knows precisely what events, facts, and the
like, they’ll have to deal with. The designs have to be suitable
for whatever might come up. Those who create an institution knowing what its remit will
be, will find it difficult, perhaps impossible to refrain from thinking about how
the design might influence the outcome in the case at hand. And achieving a desirable
balance between independence and accountability will therefore
be especially difficult. So here are some examples. Defining the terms of reference
for a special council, or a commission of inquiry. Whether the terms of
reference are broad or narrow will often have predictable
political consequences. A commission of inquiry into, with respect to some incident
of police misconduct, can be given a charge to
investigate only that incident, or to investigate what features
of the police department’s organization facilitated the incident, or to investigate a broad
class of events of which the triggering incident
was a single instance. Similarly, with a special
counsel or prosecutor charged with investigating
a corruption scandal. The charge might be to look
into the scandal alone, or into other incidents
in which those implicated in the scandal might have participated. Or into the conditions in the ministry that facilitated the
corruption, and the like. Now political actors will evaluate the proposed terms of reference, by thinking about their
political implications. Corruption investigations
offer obvious examples. At the outset, we will know who the target of the investigation is, and what the target’s
political affiliations are. The target’s allies will
want the terms of reference to be as narrow as possible,
to insulate themselves from being swept up
into the investigation. And the targets of opponents
will want the terms of reference to be as broad as possible, giving the opposite political valence to exactly the same concern. Okay, now, that’s one problem caused by not acting under
the veil of ignorance. Selection of personnel to
staff ad hoc institutions raises similar problems, where we know the partisan affiliation of potential special councils. We might ask for the United States, can a Republican Special
Counsel be impartial when investigating a Democrat? Or for that matter, when
investigating a Republican? The problem is particularly acute where we’re dealing with a
single investigator, of course. And that might suggest, incidentally, that a well-designed ad hoc institution would have multi-member leadership. Multi-member ad hoc institutions
have their own problems, though, and this is the story of the 1876 US Election Commission. In the 1876 election, there were disputed votes in three states. And if the disputes were
resolved consistently on partisan lines, Republicans would win. And yet, at the same time, you wouldn’t think that
the same kinds of problems with the vote counts occurred
in each of the three states. There were problems in each of the states, but there’s no reason to think
they were the same problem. And no obvious reason to think
that they were all skewed in favor of the Republicans or Democrats. So what happened? Needed to investigate,
you need to figure out who won the election
in these three states. Commission was designed to
achieve partisan balance in a highly charged partisan environment, because the presidency
turned on the outcome. What was the design? The Commission had five Democrats and five Republicans
appointed by Congress. They were gonna divide evenly. It had five Supreme Court justices. Two of them were identified
with the Democratic Party. That is everybody at the time thought of these justices as Democrats. And two were identified
with the Republican Party. And then the fifth one was an independent in the sense that nobody knew
what his political views were. He was going to be the tiebreaker. Name is David Davis. Basically, people expected
that Davis’s evaluation of the facts would be dispositive. Now, before the Commission began its work, Davis resigned his seat
on the Supreme Court to take a position as a
Democratic senator from Illinois. I’ll explain the politics behind that appointment in a moment. There’s now a vacancy on the Commission. Naturally, given the
design, five from the House, five from the Senate, you
were gonna have to fill it with a Supreme Court justice. But Davis was the only even
plausibly independent justice on the court, and the Commission vacancy was filled by Joseph Bradley, identified with the Republican Party. And as everybody would have expected, Bradley voted consistently
with the other Republicans on the Commission, whose ultimate decision split along party lines. Now, that particular
example may be too (mumbles) to generalize, except in a way that I’ll indicate in a moment. And here’s the basis
of the generalization. In 1876, US Senators were
chosen by state legislatures. The Illinois legislature
was controlled by Democrats. They chose Davis to be the senator. Their political calculation apparently was that he would serve on the Commission, vote with the Commission’s
Democrats out of gratitude, and then assume his Senate seat. Their political calculations
backfired, of course, when he resigned before
the Commission met. But the generalization or the
lesson of the episode then is not that personnel decisions inevitably track political alignments. But rather that political actors will try to game the selection process. And thereby will undermine
the institution’s independence and/or accountability. That is in some sense,
every personnel choice, even of nominal independence, will be viewed by partisans as potentially the product of some, perhaps hidden, impermissible, partisan, or
undesirable partisan gaming. Now, having expressed all
this kind of skepticism about the design of these
ad hoc institutions, I do want to say, particularly in Canada, that these kinds of political influences can be constrained to
some degree by norms. In Great Britain and
Canada as I understand it, strong norms dictate that
Commissions of Inquiry should almost always be headed by retired and occasionally by sitting judges. And the initial design of
the 1876 Election Commission in the US reflected a similar instinct that judges can be counted upon to be impartial to some
significant extent. And yet, of course, that very same design shows that judges might
not be impartial enough. Four of the five justices were understood to have strong party affiliations. And post legal realist world, members of the public and some lawyers will offer political descriptions,
liberal or conservative, of the track record of a judge named to head a Commission of Inquiry. The accuracy of those
descriptions may be contested, but the very fact of contention
over the descriptions introduces politics into
the public evaluation of the judge’s appointment
and performance. And here I wanna pause to give, what for me is the most either interesting or egregious recent example. Doesn’t quite fit what I’ve
said, but it’s suggestive. There’s a judge named
Sergio Moro who took charge of an anti-corruption
investigation in Brazil. And the initial subject was not narrow, it was a big corruption issue. But it dealt with only
one corrupting body, one corporation that was bribing people. But Moro followed the
tracks where they led, and greatly expanded the
scope of the investigation. Eventually, the investigation
led to the conviction of the leader of Brazil’s Workers’ Party, Luiz Inacio da Silva, and his
disqualification from office. At a time the disqualification occurred, when he was the leading
candidate for re-election, or for election in this past election. Lula’s supporters had previously claimed, based on sort of the
pattern of indictments, that Moro’s investigation was biased against the Workers’ Party. It was just based upon that pattern. But, and this is the move
that makes things interesting, in 2018, Moro resigned
his position as judge to become a minister in the
government of Jair Bolsonaro, the minister in charge with
investigating corruption. But nonetheless, a minister in this extremely conservative government. That confirmed for Lula supporters
their initial suspicions. Okay, again, it’s not quite, this is confirmation that
occurs in retrospect, whereas what I was talking about
was controversy that occurs at the appointment of
the investigator stage. However, so these are problems
with ad hoc institutions. But they may be a sensible choice when the problems they
address arise occasionally. There may be little reason to
create a permanent institution whose staff sits around waiting
for something to happen, or more troublingly, goes out aggressively to find problems to solve, simply so that the staff
has something to do. I’ve suggested in this
work on South Africa that permanent anti-corruption agencies do have an incentive to adopt broad, sometimes troublingly broad
definitions of corruption, so that there’s work for the staff to do. When they’re created,
multi-member bodies may be better than single headed ones. Ad hoc bodies may be better
than single headed ones. And other features such as norms and the public view of
judges’ impartiality may not be under
designer’s control, though. Now, on to permanent statutory IPDs. Examples from the United States include permanent legislative and
executive ethics offices, created by statutes and located
in the respective branches. Inspectors general located
within the executive departments. Examples from elsewhere
includes Ombuds offices, as originally designed. There were statutory agencies located in the legislative branch, tasked with overseeing
how executive ministries administered the law at
its point of application. And statutory election
and boundary commissions. Now, the primary concern
with respect to permanent but merely statutory
institutions is, of course, their independence from
political control or domination. By design they have
someone they report to. Ombudspersons to the legislature, inspectors general to the
department head or the president, and that person who we
should understand to be a superior officer to the institution, can control the institution’s
actions rather directly. For example, sorry, directly, for example, by removing some investigation
from its jurisdiction. Or by redefining the
scope of its jurisdiction. Or indirectly by removing
the institution’s head, or otherwise making life unpleasant for the head of the institution. Without constitutional guarantees,
statutory IPDs are always vulnerable to political control or attack. Now again, norms can offset some of this. An anti-corruption
investigator might be directed by its superior to steer
clear of some potentially sensitive matter, but might
invoke professional norms as a reason for doing so. And what we’re experiencing
in the United States is a situation where I
think the anticipation of the invocation of professional norms is restricting President Trump’s ability to control the investigation,
the Russia investigations. Legislators might tell
a Boundary Commission to take politics into
account in a substantial way, in drawing district boundaries, but the Commission’s
members might refer to professional norms that counsel them to limit the role of,
limit, not eliminate, but to limit the role of
politics in drawing boundaries. Now, because these institutions
are merely statutory, when push comes to shove,
they might lose out. But the very fact that
effectively exercising political control over a statutory IPD, will be politically contentious, might give the statutory
institution enough independence from politics to operate reasonably well. I had a conversation last
night with Professor Dawood about the Elections Canada. And I have to look into it,
but the story she told me is sort of consistent
with what I’ve just said about the vulnerability
of the institutions to political challenge. But also, the possibility
that, for a variety of reasons, that they might be in a position to resist or overcome the attempts
to gain control over them. Okay, so now I wanna turn
to the Constitutional Court as the first institution
protecting democracy, and talk a little about its relation to the other institutions as well. So, as I said, the Kelsenian
Constitutional Court was the first IPD to be
conceptualized as such. (mumbles) and Ginsburg have shown that many Constitutional Courts
have taken into themselves some of the functions of other IPDs. They resolve election
controversies, for example. Or they determine ballot qualification, in the way that electoral courts or independent election
commissions sometimes do. Brazil’s Constitutional
Court has the exclusive power to try national legislatures for crime, including corruption. And sort of the observation
here is at the moment, more than half of the
Brazilian legislators have been indicted for criminal activity. They are at the moment entitled to trials before the supreme Constitutional
Court, and sort of, more or less everybody agrees, that’s a terrible design. Bad that they’re all indicted. But given that they’re
all indicted, it’s really, you’re not gonna get much
done if you have to try these cases as original
ordinary criminal trials, before the Constitutional Court. Constitutional Court justices
are struggling to figure out some doctrinally acceptable way to say, although the Constitution says they have exclusive jurisdiction over
these kinds of criminal trials, they can figure out some way to channel some of the less significant ones for initial trial somewhere else. The trial of Lula was in a lower court under what Moro did the indictment, but that’s because he
wasn’t a sitting legislator. He had left office. Okay, so again, Constitutional
Courts have absorbed IPD functions into themselves. The reason for this
absorption, there are several. First, the one I initially
thought was most important was a simple shortage
of qualified personnel. Perhaps especially in nations
moving from authoritarianism to democracy, or from severe civil strife to a reasonably peaceful order. All these institutions
need people staffing them who are technically qualified,
and to a significant degree, independent of political control. Small nations might not have enough people who have the right
combination of qualities to staff a full range
of these institutions. Post-transition nations
might not have enough qualified people who will
be generally untainted by what they did earlier. Related, the Constitutional Court might be the only institution with
sufficient credibility to resolve election disputes,
draw district boundaries, investigate corruption, and the like. And that might be especially
true where the alternative is a completely new institution
unfamiliar to the public. The thought is getting five
to nine people to staff a Constitutional Court
might be hard enough. And once having located them, the nation’s leaders
might think it worthwhile just to put the matter aside, and assign other IPD functions
to that same group of people. Again, a sort of footnote here. A design alternative
that’s used in extremely small nations is foreign judges. Bring in somebody from outside. Hong Kong and Singapore do this. They’re not the right kind of nations. Pacific Ocean nations
interestingly use foreign judges, I think because of staffing problems. Now, having said this,
so there may be reasons to layer on the other functions into the Constitutional Court, but doing that can cause obvious problems. On the most mundane level, doing other work takes time
away from doing the central work of resolving disputes about
the Constitution’s meaning. Some IPD work may require
the use of expertise that judges trained
primarily as lawyers lack. The best examples that I know of involve anti-corruption investigations, where you have to trace money through various hands to figure out who was corrupting whom, and
what the the money flow was. In a similar vein, these
institutions, the IPDs, might have remedial
tools available to them that are more difficult for courts to use. Not impossible, but more difficult. Several constitutions include their Human Rights Commissions
in their list of IPDs. I, myself, have to think more about whether that classification is correct. But if it is, an argument
available for doing it would be that the Human Rights
Commission might find it easier than a Constitutional Court, to develop remedies for discrimination that involve ongoing
oversight, supervision, and regular revision. Now, we know from Indian
public law litigation and US public interest litigation, what Chuck Sabel calls
experimentalist litigation, that courts can do this sort of stuff, but might be easier for the
other agencies to do it. Easier and more effective to do it. The task of resolving election disputes and more generally,
supervising law of politics, implicates the Constitutional Court and what Professor Herschel
has called mega politics. And that too, might cause difficulties. The most obvious one being
that however the court resolves the controversy, one side
will think, or at least worry, that the court wasn’t impartial. Or more dramatically, that the
court’s decision was dictated not by law, but by the
judges political preferences. And that might be troublesome, not, I wanna say with respect to the
particular election dispute, but because it might
weaken the courts ability to secure compliance
with its other orders. To the extent that that ability
rests on a general sense that the court’s decisions are impartial. Now, a final question here is the relation between the Constitutional
Court and other of these IPDs. As far as I know, the
issues have not arisen. Conflicts between the courts and the IPDs haven’t arisen that often. I have little doubt but
that Constitutional Courts will routinely exercise
supervisory power over other IPDs. They will police the boundaries
of an IPD’s jurisdiction. They will assess the
procedures used by the IPDs, by looking to standards
they themselves develop. Now, that’s not an unalloyed good. Though, it’s inevitably
gonna happen, I think. I haven’t explored the question here, but one important feature of
IPDs is that their work entails the deployment of kinds of expertise, that might not be readily available to a Constitutional Court. On the jurisdiction of an
anti-corruption agency, for example, a Constitutional
Court might not understand how investigators follow the breadcrumbs through straw transactions, to identify people implicated
in a corruption episode. Even though those people
seem entirely unconnected to the investigation starting point. There’s an interesting provision in it’s either the constitution
of either Bolivia or Ecuador, that specifically charges
or gives authority to the anti-corruption
investigator to look into, I forget the phrase. I couldn’t find it quickly. Dark money and straw transactions, fake transactions designed
to conceal actual corruption. And I find the provision interesting, because the Constitution
drafters understood that they had to single this
thing out and identify it, because, I think, they weren’t sure that anti-corruption agencies
would otherwise understand that they were supposed to do this. The anti-corruption agency
can’t understand it. It’s not obvious that a
Constitutional Court would. Maybe more common, an
anti-corruption investigation might require, or at least find helpful, investigative procedures that
might seem unfair to judges not regularly engaged in
investigating corruption. In Brazil, in the Moro investigation, there was a non-standard use of plea bargaining and
coercive pretrial detention that occasioned, that judges
other than Moro, bridled at. And ultimately, Moro got away with most of what he was trying to do. But his departure as an
anti-corruption investigator, from judicial norms of perceived fairness, caused jurisdictional trouble. Now, just a little summary on this. Taking the view that Constitutional Courts will never relinquish complete control over the operation of other
of these institutions, I’ve suggested elsewhere
that the best solution to this problem is for
Constitutional Courts to say, yes, we are going to supervise them. But we should give substantial deference to the jurisdictional
and procedural choices made by other IPDs. I should note that in the
South African case study that I’ve done, the Constitutional
Court didn’t do that. They exercised very, so call
it, independent non-deferential judgment about the design of the anti-corruption agency itself, in with opinions that
were the initial opinion, sharply divided then
later almost unanimous, but that were not, in my view, a terribly persuasive as legal analyses. And it’s reasonably clear
to me that what was going on was that the Constitutional Court believed that the thing
that they were looking into, and please, almost certainly correctly, that the thing they were looking into was the strategic use of
anti-corruption investigations by one faction within the
African National Congress, against the other faction
in the National Congress. And they sided with the
people who are at the moment, the losers, and they won basically. The other faction displaced the prior one. But that’s an indication
of the kind of difficulty that actually can occur. Okay, now before concluding,
I want to address a question that’s concealed in the very
framing of my discussion. Why think of these
institutions collectively, as a new branch of government? Now part of the answer, is,
I think, simple nominalism. If you look around at
modern constitutions, designers seem to think
that these institutions have something in common,
and signal that belief by including a bunch of
them in a separate chapter or article in the constitutions. And that nominalist approach
may make sense of the fact that separate identification of IPDs is a quite recent phenomenon. Again, the South African
Constitution is 1995 or four, depending on which version
you take into account. Constitution designers now know or think that there’s something new out there, and they’ve brought that
novelty into the Constitution by emulation, rather than deliberation. Now I’ve offered a bit
more than nominalism in my account of IPDs as responding to conflicts of interest
between other institutions, with respect to protecting democracy. Unfortunately for that
functionalist approach, some institutions included
in the chapters or articles that I’ve mentioned. Don’t seem to respond to
conflict of interest problems, cultural preservation institutions
are sometimes included in the Latin, the new Latin
American constitutions. They’re lumped together with the other, the Electoral Commission or whatever. And it’s hard to see why. Maybe we could come up with an account, but I doubt that it’s motivating. So some of the things in the chapters don’t seem to respond to
conflict of interest problems. And as the example of Central Banks shows, not all conflicts of
interest do or should support the creation of one of these institutions. In the end, I leave open the possibility that we’re not dealing with
something fairly described as a new branch on the same plane as the Montesquieuian branches. Montesquieu may have been right, Sun Yat-sen may have been wrong. We may have no more
than what I referred to in the introduction as a
conjuries of institutions that have something to do with the preservation of
constitutional democracy. Some are constitutionalized in
some constitutional systems, left to statutory law. And others dealt with
through ad hoc mechanisms in the third group, and
ignored in yet other nations. Now I’m gonna conclude
by leaving the skepticism that I’ve just expressed on the table, but also by returning to the functionalist conflict of interest account
that I have sketched. Attention to conflict of
interest does help bring the function of protecting
democracy into view. I think we can see a family resemblance among numerous institutions, each one responding to a
conflict of interest problems. At the same time, we might
wonder how much actually turns on constitutional design. Perhaps I myself think probably
institutions themselves play a relatively small role in preserving constitutional democracy. I hear my example draws on scholarship on anti-corruption efforts. That scholarship makes it reasonably clear that there is essentially
no relation whatever between institutional design, and substantial reductions in corruption. A seemingly well designed institution which dramatically reduces
corruption in one nation, will have no effect in another. And a seemingly badly designed one which has no effect in one nation, may have dramatic effects in another. The relevant scholarship converges on an alternative explanation. It’s not institutional
design that matters, but political leadership that determines whether anti-corruption
efforts will succeed. So a political leader who comes to office wanting to root out corruption, and who finds a badly
designed agency at hand, is able to use that agency
to root out corruption. If somebody imports into their nation the Singaporean or Hong Kong models, which are sort of the gold standard, but doesn’t want to root out corruption, then the agencies will just not do it. So the anti-corruption example, I think, is a pretty good one based
on existing scholarship. For other institutions,
a widespread commitment to adhering to the rule of law, rather than institutional
design, might be the key. Constitution designers and
theorists can say something about institutional design. But if design doesn’t matter much, maybe we should see the domain
of concern over preserving democracy to political
scientists and sociologists. I, myself, think that
wouldn’t be a bad outcome. We may just not be good at, or the things that we know about may not have much to do
with preserving democracy. And the things they know about, may actually have to do
with preserving democracy. So we might wonder whether IPDs can actually effectively
protect democracy. And I’m gonna end by posing that question about the first institution for protection of constitutional democracy,
the Constitutional Court. Samuel Issacharoff has argued
that Constitutional Courts might be able to deploy their resources to stabilize fragile democracies. And I think by implication, might be able to protect
democracies from collapsing. Now, I confess that I read
his book as identifying almost universally
unrealized possibilities. He does identify capacities
that these courts have to stabilize or preserve democracy. They might have the
theoretical ability to do that, but experience strongly suggests that it almost never actually occurs. I have to say, and this
is what I’ll end on, that doesn’t mean that
Constitutional Courts and perhaps other IPDs
don’t do some valuable work, but it might be that protecting democracy isn’t among the valuable things they do. Okay, thank you. As I said, I’m happy
to entertain questions, comments, and people
telling me where I’m wrong, and what other things I should look into, because this is the beginning. Not the beginning, it’s
sort of the second version, or the second step in what is
going to be a larger project. And so I am completely open to thinking about what else I should do. Yes. MAN: To begin with, sir, very
interesting presentation. And my comment or theory is
that of the human factor. And the example I’m going to
use, the most recent election in the Democratic Republic of Congo, where there is a Constitutional Court designed in terms of norms,
as to what you suggested in terms of representing the will, democracy, the values, the norm, all that. How do you deal with that, when you’re dealing with a
gangster, despotic regime? And I have met with several of the judges during my time in Congo,
who live in absolute fear. And some are put in for the simple reason to do as they are told, and
life will be good for them. Because the election in
Congo, shall we say politely, was totally rigged, flawed. How do you deal with the human factor? I’ve done some work on what I call authoritarian
constitutionalism. Lots of the things that we
value about constitutionalism are extremely difficult to preserve in authoritarian or even
dominant party systems. And so, the answer may
well be the only thing to do about that is to get rid
of the authoritarian system. There’s actually one alternative which South Africa
illustrates, interestingly, and Singapore in its own way, does too. Excuse me, so in South
Africa, as I mentioned, you have the dominant
African National Congress. The ANC has won every
election and is likely to win elections into the foreseeable future. But within the ANC there are
reasonably stable factions which operate within the
ANC as competitive parties. And that has some consequences. I have no idea about Congo
or whether it’s possible. But in your kind of example,
maybe the thing to do is to encourage the existence
or creation of factions within the authoritarian’s ruling party. Sort of footnote here, what
I’ve articulated about factions within the party is to some extent, the official theory of the
Chinese Communist Party. The reality doesn’t match very well because they have been, except for brief periods, intolerant of permanent
factions within the party. But the theory of democratic
centralism is that we can build these mechanisms
into the party structure. That’s all. Again, maybe the answer
is constitutional design can’t deal with the kinds
of problems you’ve raised. Yeah. WOMAN: Thank you so much for your lecture. It was extremely interesting,
especially for me. I’m from Brazil, and I
actually spent the entirety of last year writing a paper
about the (mumbles) case. So the comments about
(mumbles) are very well taken. My question to you is who
is gonna create this IPD? I think Brazil might illustrate a little bit of the challenge here. Since 50% of the legislature is indicted, they don’t have (mumbles) to
be the ones making the reforms. Right? So there is an idea that
there is gonna be some kind of legislative change for
these to be put in place, but if the legislators
don’t have any incentives to be accountable, then
you create sort of a track where the entire system is just rigged, and there is no way out of it. And if they mind, I just
wanna add a second question, especially to hear your
take about Sergio Moro. I think the now confirmed
suspicions that there were political motivations behind
the way he was acting, are well taken, and I’m, by no
means, trying to defend him. But one potential interpretation
of what he was trying to do was within the confines of a system that were not going to reform itself, he was stretching the
law as much as he could to create accountability, considering that Brazil
had not sentenced anybody for corruption since
(mumbles) in the ’80s, until one particular case
before (speaking off microphone) you’re probably aware of. So that’s one potential
interpretation of what he’s doing. Whether he was doing for
political reasons or not, I think that’s a valid
separate question, right? Could that be thinking as a strategy, considering that you might find yourself in a system (speaking off microphone)? Yeah, so let me address both parts of your very interesting question. The first part is how do these. I actually deliberately talked about constitutional designers
without talking about what incentives actual
constitutional designers have when the moment of design occurs. I think the only thing I
talked about relevant here was the behind the veil kind of issue. There are real problems in this, call it, rationalistic approach
to constitutional design. My favorite formulation
comes from Chuck Sabel writing about other related,
but different, issues. And he says that you get this kind of design thinking, occurs when everybody knows,
literally everybody knows, that things are going really badly. And everybody agrees that
the things that they’ve tried so far haven’t helped. In that situation Sabel says, people might be able to step back and try to figure out
what’s going to work. Now those conditions are quite stringent. Even constitutional design
in time zero situations, post-conflict, transition
from authoritarian, there are politics associated with the constitutional
design at that moment. It’s not entirely a
rationalistic enterprise. And I can say, I have no handle on that, at least in this project. The second point is your point
about Moro’s expansiveness. And this is something that I say briefly in the South African piece, but will expand on as I revise it. Your account of Moro’s,
call it motivation, might well be correct. The problem is that he’s a lawyer. I mean, I will assert this as a fact. His behavior reflected
a failure to understand how the kind of investigation
he was pursuing, was going to bring about
or exacerbate the collapse of the party system in Brazil,
which produces Bolsonaro. The party system in Brazil
is highly fractionated, and personalized, and so on. And when you do an
aggressive investigation in that circumstance, it’s different from doing
an aggressive investigation in a nation with a stable, two,
three or four party system. Because he was a lawyer and
not sensitive to politics, he just messed things up. And so, just go back to the
design of these institutions, my claim is that when you’re designing an anti-corruption agency, you want the head or multi-member heads actually to know something about politics. Some part of the Brazilian investigation involved payments elicited by the, I think it was the treasurer
of the Workers’ Party. That went primarily to
building the Workers’ Party as an institution, not to the personal
benefit of the treasurer. Some of them may have done that, but mostly this was corruption. It was corruption in the
sense that the treasurer said, if you donate X to the party, I will guarantee that you will
get a construction contract for an amount greater
than would be awarded in a competitive bidding process. So that is corrupt. But it’s not obvious that
in a world where parties are as weak as they are in Brazil, it’s something you should worry about. Maybe you should. Okay, that’s the problem. So, okay, I’ll stop with that. Anything else? Okay, well, thank you. (audience applauds)

Leave a Reply

Your email address will not be published. Required fields are marked *