Why Democracy Doesn't Work
Why Democracy Doesn't Work
INTRODUCTION
“Democracy is the worst form of government, except for all the others”, said Winston
Churchill in 1947. His quote indicates our reality: a not nearly perfect political system,
but the best we could possibly have at the moment. Sure, “power to the people” sounds
great, but is this actually good? Another quote from Churchill suggests a negative
answer, “The best argument against democracy is a five-minute conversation with the
average voter.” Democracy might be the best system out there, among our current
alternatives, but that doesn’t entail that it’s a good, desirable system. I intend to show
why.
It’s interesting to note that some theories of justice are bound to some concrete forms of
government. Therefore, we can see how Rawls’ first principle of justice, the Equal
Liberty principle, binds him to Democracy. By this principle, society is so structured
that all persons hold the same political liberties: everyone has the right to hold public
office and to affect the outcome of national elections. Then, it’s difficult to see how
Rawls’ sense of justice could be upheld in a non-democratic society, where some people
would be stripped of their political liberties.
But we needn’t be bound to the pursuit of democracy. In Anarchy, State and Utopia
(ASU, for short), the libertarian Robert Nozick chases after the minimal state, a “night-
watchman state” whose only function is the protection of its citizens’ rights against
violence and theft. In addition to an explicit critique of democracy in the chapter
“Demoktesis”, Nozick’s idea of the minimal state implicitly carries an incompatibility
with a democratic system, as there is no public law, and no actual body politic that
participates in decision and policy making. Anything more than this minimal state
would actually be coercive and unjust, all our democratic states too.
Furthermore, democracy is deeply flawed. In his recent book Against Democracy, Jason
Brennan, also a libertarian, argues that democracy should be judged by its results, and
these don’t look promising. As an instrumentalist, Brennan is more than ready to
abandon democracy for any other alternative that works better.
For the purposes of showing democracy’s failure, I will first turn to Brennan and his
profound argument against it. The next step is seeing Brennan’s own proposal:
epistocracy, democracy’s substitute.
Next, I will follow Nozick’s reflection on Utopia. In the third and final part of ASU,
Nozick argues that Utopia, the imagined community any of us would want to inhabit,
actually consists in a framework for building utopias, experimental communities that
1
we could build and structure conjointly, in an effort to create societies its inhabitants
would truly want to occupy. This philosophical digression is an abstract, albeit
interesting tool by which Brennan’s proposal for epistocracy gains new force and
justification.
1. AGAINST DEMOCRACY
In a new preface to his book, Against Democracy, Brennan says the following:
“Democracy is a tool, nothing more. If we can find a better tool, we should feel free
to use it. Indeed, […] we have a duty to use it. Justice is justice. Bad decisions are
not rendered just simply by political fiat. Political decisions are high stakes. How
dare anyone make such decisions incompetently?” (Brennan, 2017, p. xiv).
In this short quote lay the key elements of Brennan’s argument against democracy:
1) Democracy as a tool.
2) Political competence.
The core of Brennan’s argument against democracy is founded on what he calls the
“competence principle”, which states that “every individual high-stakes political
decision ought to be made competently and in good faith by what is generally a
competent decision-making body” (Ibid, p. 160). We have a right to a competent
government, as he argues in chapter 6 (“The Right to a Competent Government”).
Therefore, if it’s the case that political decisions made through democracy are not
2
competent decisions we must reject democracy on the whole. Similarly, he makes his
case for epistocracy, basing it upon an “antiauthority tenet”:
First, Brennan goes on to counter the “epistemic” defence of democracy. This defence
tries to show how competence is an emergent feature of democratic decision-making;
the electorate tends to produce smart decisions as a collective, overriding individual
incompetence. This defence is possible through three mathematical theorems: the
“miracle of aggregation theorem”, “Condorcet’s jury theorem” and the “Horn-Page
theorem”. This a priori defence of democracy fails, as Brennan responds to each
possibility, showing that these theorems do not apply to real-life democracy and that the
3
empirical evidence mentioned above gives us more than good reasons to doubt
democracy.
The second objection Brennan faces rests on empirical grounds, arguing that, while
individual voters are incompetent, democratic governments are not. This would explain
why democracies have typically performed better than other regimes. Such an objection
can have many different forms, which Brennan faces.
One possibility is that party systems “reduce the epistemic burdens of voting” (Ibid,
p. 195); voters have a “cognitive shortcut”, they can treat all candidates of a party
as a homogenous group, and thus can act as if they appeared to be well-informed.
Brennan gives four reasons why this isn’t enough. First, studies indicate that many
voters don’t even know what policies different political parties defend. Second, to
vote competently “requires tremendous social scientific knowledge” (Ibid, p. 196),
such as the possible consequences, good or bad, of their policies; it’s not enough to
know the candidate’s preferences. Third, as we saw, voters tend to be hooligans,
consuming information in highly biased ways. Fourth, as Both Ilya Somin and
Brennan argue, “the quality of the candidates […] depends in large part on the
quality of the electorate”, and as voters are extremely ignorant and misinformed,
the quality of the candidates is consequently low.
Another possibility is that, as Gilens claims, the median voter theorem is wrong,
and politicians aren’t representing the position of the median voter, but are actually
responding to the preferences of richer voters. There might be an upside to this,
given that empirical data shows that higher income voters are also high-information
voters. Brennan happily accepts this as it only goes to show that “democracy works
better than it otherwise would, because it doesn’t exactly work” (Ibid, p. 198). In
fact, this might even constitute a point in favour of epistocracy; it shows that
governments work best when responding to the preferences of the best informed.
A final possibility is that, while the voting public is largely irrational and
misinformed, it actually doesn’t get what it wants. According to Brennan, there
might be “a wide range of political bodies and administrative procedures that
mediate between what the majority of the moment appears to want during the
election and what laws and rules actually get passed” (Ibid, p. 199). This could
mean that there are sufficient checks in place in democracy so as to make electoral
decisions low-stakes political decisions. If this were the case, the competence
principle would not apply, as voting wouldn’t be sufficiently high-stakes and
government itself is competent. Brennan doesn’t buy this possibility. Major
elections “remain high stakes” (Ibid, p. 202). While elections don’t directly
determine what policies get passed, they do have a major influence in the
probabilities regarding which will eventually get implemented.
4
3) Conclusion: epistocracy
In light of everything said so far, we have reasons to reject democracy. At its base, the
voters, democratic decision-making is not handled competently. There is enough
empirical evidence that points to the conclusion that voters are irrational, misinformed
and biased beings that are not sufficiently prudent nor wise and therefore should not be
trusted with deciding political matters. Their incompetent decisions are then imposed on
innocent people, left to suffer the consequences. By the competence principle, we have
a right to be protected from this incompetence. And while the government itself might
act competently, at least most of the time, this doesn’t justify keeping democracy alive.
Less so when, thinking instrumentally along with Brennan, we appear to have a better
alternative: epistocracy.
Epistocracy, at its roots, means “rule of the knowledgeable”, much like what Plato
espoused with the rule of the philosopher-king. An epistocratic political system, then,
formally distributes power “according to competence, skill, and the good faith to act on
that skill” (Ibid, p. 14). Brennan’s proposal of epistocracy is a modest one, as he himself
admits that “Realistically, epistocracies will still feature the rule of hooligans rather than
vulcans, although epistocratic hooligans may be more vulcan-like than in democracy”
(Ibid, p. 207).
Epistocracies will generally share the same institutions as democracy. The main
difference is that not everyone has, by default, an equal right to vote or run for office.
Out of many possible forms of epistocracy (some propose unequally distributing votes,
others leaving voting rights/restrictions to chance) Brennan favours one: government by
simulated oracle. Drawing on the character of Pythia the Oracle, who is wiser than all of
us put together, and to whom we should always defer to for decision-making, Brennan
points to the possibility of simulating such an oracle. In chapters 2 and 7, Brennan had
drawn on social scientist Althaus’ empirical work on the relation between citizens’ level
of information and their policy preferences. Althaus had shown that low-information
voters “have systematically different preferences from well-informed ones, even after
correcting for the influence of demographic factors such as race, income and gender”
(Ibid, p. 34). Through this “enlightened preferences method”, we can estimate the
electorate’s policy preferences if they were better informed. This can ultimately be
applied to elections, as Brennan shows:
“Suppose there is a range of candidates from various political parties. We can ask
citizens to provide their anonymously coded demographic information and then take
a test of basic objective political knowledge. Then they rank the candidates from
most to least favored. Using these data, we can determine how the public would rank
the candidates if the public were fully informed. Whatever candidates ranks the
highest, wins” (Ibid, p. 222).
5
2. NOZICK ON UTOPIA
In an effort to shed light on the possible shortcomings of his minimal state, Nozick
takes up the investigation of utopia, as his minimal state might not invoke inspiration
like the idealisations of utopia do. While I have no interest in the minimal state itself,
Nozick’s inquiry into utopia provides me with a philosophical tool with which to further
Brennan’s criticism of democracy and his proposal for epistocracy.
At this point, the model for utopia consists in imagined possible worlds (with some
constraints that Nozick introduces, which don’t interest me at present). How do we
project this model to our world? In actuality, possible worlds correspond to a “diverse
range of communities, which people can enter if they are admitted, leave if they wish to,
shape accordingly to their wishes” (Ibid, p. 307). Given the divergence between real life
and imagined possible worlds, many of the conditions we could satisfy in the latter
can’t actually be satisfied, such as the limited number of possible communities, or the
impossibility of finding your every value satisfied in any one community. Nevertheless,
Nozick remains optimistic and states that we can “satisfy many of them”.
“Utopia”, therefore, “is a framework for utopias, a place where people are at liberty to
join together voluntarily to pursue and attempt to realize their own vision of the good
life in the ideal community but where no one can impose his own utopian vision upon
others” (Ibid, p. 312). Because people are different and complex, plurality is the
condition of utopia. Finally, utopia is “the environment in which utopian experiments
may be tried out” (Idem).
How is any of this relevant for Brennan’s substitution of democracy with epistocracy?
As Brennan himself notes, “Whether we should prefer epistocracy to democracy is in
part an empirical question, which I am not fully able to answer. […] There are good
reasons to think epistocracy would produce better results than democracy with universal
6
suffrage, yet there are reasons to worry it will not” (Brennan, 2017, p. 228). He then
explains how the English politician Edmund Burke, reflecting on the disasters brought
on by the French Revolution, wrote that humans are not smart enough to rebuild society
through revolution, beginning anew. The French Revolution showed that institutions
“that seem unjust on philosophical reflection turn out to serve useful purpose” (Ibid, p.
229). This all led to “Burkean conservativism”, the idea that “we must be extremely
cautious when making radical changes to existing institutions. […] Experimentation
with new forms of government is dangerous” (Idem).
If all this is true, Brennan might find in Nozick a justification for epistocracy. In
philosophising about utopia, Nozick paves the way for new, innovative proposals (such
as Brennan’s) that might bring us closer to a community which all would want to
inhabit. Democracy, as Brennan has effortfully shown, is unjust; epistocracy might
relieve us from these injustices. How dare anyone deny us our right to strive for utopia?
It is then time to experiment. I shall try to see how far epistocracy can take us, if we
commit ourselves to the effort of trying to arrange our institutions so as to achieve
justice.
This is not the type of epistocracy I wish to experiment with. I want to take Brennan’s
government through simulated oracle to the next step, over and beyond elections. I think
7
simulated oracle can work, not just for choosing the best-suited political candidate, but
also for choosing and implementing actual policies, actual laws. If by enlightened
preference method we can cross-reference citizen’s degree of knowledge with their
political inclinations to figure out what policies would be preferred in a situation of high
information, why not use it to actually choose and implement these policies? Through
such empirical study, we can skip political representation altogether. Why choose to
have parties struggling for our support in an electoral feud, which might then not fully
represent us in parliament when choosing what laws to pass, when we can discover the
actual political decisions citizens with an ideal degree of knowledge would make?
Brennan has made sure that knowledge is correlated with good decision-making in
politics. He has given ample evidence that the political preferences of high-information
voters seem to be good preferences. He says, for instance, that high-degree voters all
tend to prefer economic policies which economists themselves tend to agree over, such
as less government intervention and control over the economy. They are also
systematically more liberal when it comes to social issues: more pro-choice, less
punitive crime measures, less military intervention, more affirmative action, etc. If the
people know what they want, and we now know that some wants are better than others,
and that we can actually determine these better wants, I don’t see where the mystery is.
Let’s suppose, for the sake of argument, that government by simulated oracle as I have
construed it, post-elections, can be implemented. Where does this leave politics, as we
now understand it?
Let’s take a basic political institution, parliament or congress, for instance. In most
democratic governments (parliamentary governments), there is such an institution
where policies and laws are discussed by deputies, and finally passed (or discarded). In
such places, elected representatives exercise the legislative power of government. Laws
are voted for by these political characters, who act in name of their political party and,
ultimately, for the people. Nevertheless, this is mere representation. Citizens don’t
really get a vote in what laws are to be passed. They merely get to choose what parties
can have a greater influence in this process.
Through epistocracy, as I propose it, there is no need for such a discussion regarding
laws. Parliament is no longer the home of representatives who choose policies after
“deliberation” and discussion. With simulated oracle, parliament would be “occupied”
by social and political scientists, who use their methods (such as the enlightened
preferences method) and devise newer, better ones, to figure out what policies would be
preferred in ideal knowledge conditions. Such an epistocracy closely resembles a
technocracy. It seems that this is truly the rule of the knowledgeable. No longer would
we depend on contingent discussions, by characters that, as Brennan argues, are of low
quality, incompetent, biased, and corrupted, in our search for justice.
8
4. CONCLUSIONS
I set out to show the faults of democracy, a system that is deeply rooted in our society,
but deeply flawed.
A further line of analysis, now, would take up the question of how to implement such a
system. I truly think the social sciences should shift their attention to these questions.
This kind of experimentation really is, as Nozick said, the way to utopia. And what can
be more important than giving our all to the pursuit of justice, to building a community
for all?
9
BIBLIOGRAPHY
NOZICK, Robert. Anarchy, State, and Utopia. New York, Basic Books Inc (1974)
10
CONTENTS
INTRODUCTION 1
AGAINST DEMOCRACY 2
NOZICK ON UTOPIA 6
CONCLUSIONS 9
BIBLIOGRAPHY 10
11