Is (limited or moderate) epistocracy better than democracy?

Abstract: This discussion critiques moderate epistocracy by focusing on the demographic objection i.e the concern that electorates which are chosen for political competence may have epistemically damaging features. Restricting suffrage in this way leads to injustice but also potentially worse results. This is due to several reasons, including that this electorate may have an information gap making them unable to advocate for those disenfranchised, but also because a more homogenous electorate may lead to a lack of cognitive diversity. In turn this may cause greater problems for wider society in the face of uncertainty and creates potential fragility. Hence, the case for epistocracies is neither morally convincing nor instrumentally strong.

Introduction

Epistocracy is a form of government in which it is believed the wiser and more competent should rule. Classic epistocracies may be characterised with concepts like Plato’s philosopher kings but this is of course a fictional ideal; more realistically, is the notion of limited or moderate epistocracies. Moderate epistocracies are motivated in two ways: mainly that it is presumed they would reach better outcomes but also they are less unjust than democracies due to democracy’s violation of the competence principle (Brennan 2011). This discussion pays particular interest in the demographic objection which notes that an electorate which is tested for political competence may result in a group of voters with potentially damaging characteristics. From this we may argue that the demographic objection illustrates not only greater injustice but also potentially suboptimal outcomes compared to democracy.

The Case for Moderate Epistocracy

Epistocracies can take different forms, but they generally begin with the assumption that a significant portion of voters in a democracy are politically ignorant (Brennan, 2011). This is not an unreasonable assumption; in studies conducted, most people are poorly informed on basic facts from their local politicians to estimates of national spending (Brennan 2011). As a consequence of voter ignorance, epistocrats implies the electorate will often vote ineffectively. Naturally, the moderate epistocratic argument suggests if more politically competent votes made up the electorate this would lead to better policy outcomes. The methods of achieving this more politically competent electorate varies. For instance, John Stuart Mill (1861) advocated for plural voting in which the more educated would get more votes than their uneducated counterparts. Meanwhile other forms of epistocracy have different ways to avoid “polluting the polls” (Brennan, 2009, p.535). For example, restricted suffrage sees votes being given to those who can prove themselves sufficiently well informed (such as through an exam), whilst preventing the incompetent from voting (Brennan, 2011, p.714). Hence, by limiting the dilution of competent voters in the electorate it is implied that epistocracy surpasses democracy, at least in terms of providing better outcomes.

Additionally, for some like Brennan (2011), moderate epistocracies are not just instrumentally better than democracy, but through the competency principle they are morally superior too. Under the competency principle, Brennan (2011, p.704) suggests it is unjust to ‘deprive citizens of life, liberty or property, or to alter their life prospects significantly’ as a result of incompetent rationale alone. To reinforce this intuition, consider a jury that comes to a verdict through unreasonable means. Brennan gives at least three ways a jury can come to questionable conclusions: firstly, they may be ignorant and refuse to listen to evidence; secondly, they may be incompetent and come to blatantly inaccurate conclusions; and finally come to a verdict based on morally unacceptable reasons e.g flagrant or extreme prejudice (Brennan, 2011, p.703). In these kinds of cases it is natural to be tempted to disregard the jury’s conclusions due to the methods which led them to that particular verdict. Similarly, one may argue if an electorate is made up of mostly voters who come to conclusions in these ways their votes should be seen as morally questionable (Brennan, 2011, p.710). To clarify, this does not suggest authority only has legitimacy when they are correct, but rather they lack legitimacy if they reach conclusions in unacceptable ways. (Brennan, 2011 p.705). Additionally, Brennan notes that while restricting suffrage may be a violation of the qualified acceptability requirement, i.e the basis for the distribution of political authority must be accepted by all qualified points of view (Estlund, 2008, p.33), he ultimately concludes this is less unjust than democracy’s violation of the competency principle. Therefore, epistocracies may not only lead to better policy decisions from the electorate but that it is morally superior to democracy as well.

The Demographic Objection and Disadvantage Groups

While we have discussed that an epistocratic electorate may have biases which actively disfavour certain groups, we may argue that even if the electoral demographic was well intentioned they may still lack the necessary information to advocate for those outside of this demographic. Mendez (2021, p.154) calls this the information gap problem. This is not a particularly new idea, Du Bois (1920, p.146) argues, even ‘[w]ith the best will and knowledge, no man can know women's wants as well as women themselves. To disfranchise women is deliberately to turn from knowledge and grope in ignorance’. In this way, epistocracy unjustly and functionally fails to capture the desire of those in groups unable to qualify for suffrage. Indeed, Landemore makes the point of noting ‘how ironic would it be, in a moment like this, to silence the very voices that our polity needs the most’ (Landemore, 2021, p.206). To further Mendez’s information gap problem, Mararo (2018) points out that there are asymmetries between potential demographic groups on the same issues. These asymmetries act as further separation in understanding between the qualified electorate and non-qualified electorate. For instance, Mararo (2018, p.213) argues the effect of ‘[e]conomic policies will have on citizens who are better-off compared to those who disadvantaged citizens have more to lose from bad go their affluent fellows’. Hence, an electorate deemed competent may take on more risk in policy decisions since the potential risk in terms of proportional harm for them is far lower. In turn, the demographic objection shows epistocracy has potential to be susceptible to selecting an electorate that unjustly fails to advocate for disadvantaged groups whether this be through bias or a lack of information. Moderate epistocrats may want to argue that those disenfranchised by a competency test would not be able to vote in ways that promoted their own interests anyway. Even if this was the case, as illustrated previously, it is evident that supposedly competent voters would not be able to aid the disenfranchised either and to exclude them from the electorate would be an unnecessary injustice.

The Demographic Objection and Wider Society

It seems then the demographic objection shows potential pit-falls in epistocracies for those in disadvantaged groups, but these issues extend to wider society as limiting the electorate does more harm than good for decision making. The slogan, ‘one man, one vote’ reiterates a core principle of democracy, i.e it proposes everyone has equal political contribution (at least in the de jure sense). For Landemore this is an invaluable quality of democracy, not simply because it may be the morally correct action but because electoral inclusivity ‘is the most parsimonious way to get us the cognitive diversity we need’ (Landemore, 2021, p.203). Having cognitive diversity allows for a much greater deliberation process and thus better results. Additionally, Landemore’s claims are supported by Hong and Page (2004) who contend that a diverse group of agents would outperform teams made up of the best-performing agents in problem-solving processes. Thus, the demographic objection shows that in having a more homogenous electorate, the assumption that epistocracies would lead to better outcomes is weakened.

While Brennan attempts to argue restricted suffrage epistocracies might make better decisions, it seems that assumption is not necessarily true, this is reinforced when epistocracies face extreme uncertainty. Advocates for epistocracy misunderstand that in maximising the electorate, democracy yields results which are robust – unlike epistocracy which potentially encourages fragile outcomes, even if the average voter competence is lowered. Consider the increasingly technocratic state, Australia, where in 2019, wildfires took thousands of homes and the lives of billions of animals. In the aftermath, some argued that less damage would have taken place had Australia not criminalised and dismissed the Aboriginal practice of preventative bush burnings, (Landemore, 2021, p.207). Landemore uses this example in part because it is not difficult to imagine the Aboriginal people being excluded by political literacy exams in which Brennan’s (2011) restricted suffrage may rely on. Hence, it demonstrates the potential fragility that occurs when voices are disregarded, as supposedly competent people make decisions which they hubristically believe to be for the greater good, only to find out they may have made a mistake.

To further highlight the benefits of diversity in the face of uncertainty, consider another example but from finance. While markets are known for being difficult to predict, Harry Markowitz won a Nobel Prize in economics for developing an elaborate investment model to which one could calculate future risk if one knows future uncertainty (Taleb 2005, p.188). Yet for Markowitz’s own retirement fund he opted for a heuristic approach by instead distributing his wealth evenly across all options (Gigerenzer, 2014, p.22). Indeed, work done by DeMiguel, Garlappi and Uppal (1990) confirms that this heuristic approach is a far more robust portfolio model than Markowtiz’s nobel-winning model was. As Landemore (2014, p.169) notes, Markowitz reluctance to use his own model ‘is simply the realization that the actual world is much more unpredictable than the world modeled by economists’. To conclude then, if an epistocratic electorate may not even be able to effectively advocate for those unable to pass such competency tests. It is even more optimistic and perhaps impossible to believe an epistocratic electorate could predict “which voices will matter or not in the future based on some political literacy tests designed by social scientists.” (Landemore, 2021, p.204). As such, epistocracies seem potentially more fragile in the long term and the assumption that allegedly more voters would produce better policy decisions is not necessarily so straightforward due to the lack of cognitive diversity which democracies suffer less from.

Understandably, Brennan (2011, p.723) suggests this kind of argument is simply ‘Burkean conservativism’. Yet this does not diminish Landemore’s (2021, p.207) concerns that we ought to remain cautious and understand that “[p]olitical knowledge is a living thing and comes in various guises. We should be humble about who has it and about the possibility of defining or identifying it ex ante.” In this sense, attempting to introduce epistocracy especially at scale seems like a precarious venture, particularly if we have reason to believe it would not necessarily lead to better outcomes due to a lack of cognitive diversity.

Does Jeffery’s limited epistocracy prevent any of these issues?

So far we have discussed that Brennan’s restricted suffrage has several flaws, but Jeffery (2018) offers a different kind of epistocracy in what she calls a limited epistocracy. This is defined as ‘qualified rule by expert institutions’ (Jeffery 2018, p.412) in which such institutions issue directives that give the electorate multiple actionable options. The issue with this is obvious, and one the demographic objection highlights; for clarity, it may be necessary to spend fortunes to be sufficiently educated to enter one of these institutions which in turn leads to certain groups having less political authority. Jeffery (2018, p.429) admits to underrepresentation, highlighting that in 2013 in the U.S, 51% of those employed in engineering and science were white males, as compared with 20% white women, 3% black men and 2% black women. In order to solve this Jefferys (2018, p.430) posits that underrepresented ‘groups need to be guaranteed some latitude in their deliberation and decision-making that doesn’t conflict with their reasonably held commitments and values’. Indeed, if one could guarantee that such groups were not limited by the options provided by those institutions, this principle might suffice, but it is unclear if this could be thoroughly upheld. Hence Jeffery’s limited epistocracy remains questionable compared to democracy which guarantees deliberation.

Conclusion

To conclude, the case for epistocracy is two. Firstly there is a presumption that a more politically competent electorate would vote more effectively for policy decisions and thus have better outcomes. Secondly, under the competence principle, epistocracy is less unjust than democracy. While these are compelling arguments, the demographic objection arguably deals with both. Estlund (2008) demonstrates with the demographic objection that there is dangerous potential for an electorate to be selected in which they share epistemically damaging features. Even so, if we assume this electorate is well intentioned, there is no guarantee that they will be able to meaningfully identify the issues that the disenfranchised face. As a result this disregard is arguably more unjust than the violation of the competency principle. Furthermore, the demographic objection shows that a more homogenous electorate leads to a lack of cognitive diversity. This is a great concern for Landemore (2014) who suggests it may lead to unforeseen consequences, whereas democracy through inclusive principles generates outcomes which are more robust. Therefore, epistocracy is not better than democracy on these accounts.

Bibliography

• Brennan, J. (2011). The ethics of voting. Princeton: Princeton University Press.
• Brennan, J. (2016). Against democracy. Princeton: Princeton University Press.
• DeMiguel, V., Garlappi, L. and Uppal, R. (2009) ‘Optimal versus naive diversification: How inefficient is the 1/N portfolio strategy?’, The Review of Financial Studies, 22(5), pp.1915–1953. 
• Du Bois, W.E.B. (1920) Darkwater: Voices from within the veil. New York: Harcourt, Brace and Howe.
• Estlund, D. (2008). Democratic authority: A philosophical framework. Princeton: Princeton University Press.
• Gigerenzer, G. (2014) Risk savvy: How to make good decisions. London: Viking.
• Hong, L. and Page, S.E. (2004). Groups of diverse problem solvers can outperform groups of high-ability problem solvers. Proceedings of the National Academy of Sciences, 101(46), pp.16385–16389.
• Jeffrey, A. (2018) ‘Limited epistocracy and political inclusion’, Episteme, 15(4), pp.412–432. doi:10.1017/epi.2017.8
• Landemore, H. (2014) ‘Democracy as heuristic: The ecological rationality of political equality’, The Good Society, 23(2), pp.160–178.
• Landemore, H. (2021) ‘Against Epistocracies’, in: Debating democracy: Do we need more or less? New York: Oxford University Press. Online edition, Oxford Academic, 23 December 2021.
• Méndez, M.P. (2022) ‘An epistemic problem for epistocracy’, Social Epistemology, 36(2), pp.153–166. doi:10.1080/02691728.2021.1992531
• Mill, J.S., 1861. Considerations on representative government. London: Parker, Son, and Bourn.
• Moraro, P. (2018) ‘Against epistocracy’, Social Theory and Practice, 44(2), pp.199–216. 
• Taleb, N.N. (2005) Fooled by randomness: The hidden role of chance in life and in the markets. 2nd edn. New York: Random House.