Oxford Uehiro Prize in Practical Ethics: Why Don’t We Just Let The Wise Rule?!

This article received an honourable mention in the undergraduate category of the 2022 National Oxford Uehiro Prize in Practical Ethics

Written by Alexander Scoby, University of Cambridge

Throughout history, democracy has been accused of producing objectively sub-optimal outcomes because it gives voice to the ‘mob’. 1 Recently, Brexit and the election of Trump have been the favoured examples.2

The supposedly poor epistemic performance of democracy has served as a springboard for epistocracy, loosely defined as any political arrangement where the ‘wise’ (or competent) have disproportionate political authority relative to the rest of the population.3

I argue that against a background of structural inequality, an epistocracy is unlikely to epistemically outperform democracy. By doing so, I hope to undermine the appeal of epistocracy and ‘defend’ democracy from a competitor.

My essay has six sections. In Section I, I introduce the ‘epistocratic argument’. In Sections II-V, I offer three related, mutually reinforcing epistemic objections to epistocracy grounded in an understanding of structural inequality. In Section VI, I consider the implications of my answer.

  1. The Epistocratic Argument
    Epistocratic Argument (EA):
  2. Epistocracy would tend to produce better outcomes than democracy.
  3. If epistocracy would tend to produce better outcomes than democracy, we ought to prefer epistocracy to democracy.
  4. Therefore, we ought to prefer epistocracy to democracy.

Before assessing the EA, it is worth noting that it assumes objectivism and cognitivism with respect to claims about political outcomes. That is, it assumes that some outcomes are objectively better than others and, further, that we can know the objective value of outcomes. This assumption is justifiable, though not uncontroversial. 4 I accept the assumption in this essay, since both sides in the debate do.5

A less plausible assumption in the EA underlies premise (2): epistemic performance is all that matters when assessing political systems against one another. This is not true. In fact, arguments against epistocracy often invoke normative considerations about procedural fairness or equality. 6 In this essay, however, I grant (2). This is purely for dialectical value: by engaging with the epistocrat on their assumption that instrumental value is all that matters in a political system, my argument is stronger than it would be if I relied on normative considerations that they might not share.

(1) is the interesting premise for me. I argue against its truth in Sections II-V. However, there are two reasons which might initially make it convincing.

First, ‘The Value of Wisdom’. Generally, it is reasonable to expect that wisdom is positively correlated with a tendency to produce better outcomes. For example, the legally wise lawyer will tend to produce better legal outcomes for a client than a novice would. The same is true in domains as diverse as cooking and football. In these domains, then, it is intuitive that giving these experts disproportionate authority will foster a tendency to produce better outcomes. We might reasonably expect that politics is similar. If so, (1) is plausible.

Second, ‘The Burden of Ignorance’. Empirically, many voters are ignorant of relevant, important facts. 7 Generally, ignorance (or a lack of wisdom) hinders a tendency to produce the best possible outcomes; this is why the legal novice will rarely produce the best legal outcomes. Epistocracy, by relatively or absolutely disenfranchising the ‘unwise’, is better placed to mitigate the influence of the ignorant in decision-making than democracy, where everybody theoretically has equal political authority. In turn, this lack of ‘burden’ supports (1).

(1), then, is not implausible. Nonetheless, it is false, at least in our non-ideal world.

  1. Structural Inequality

My rejection of epistocracy depends on a background understanding of structural inequality, its existence in current society, and epistocracy’s essential maintenance of it.

By structural inequality, I refer broadly to inequality produced by social or political structures, where ‘structures’ are understood as encompassing practices and institutions. Structural inequality can take many forms – material, social, political.8

That structural inequality currently exists will not be debated here but justifiably assumed, since its existence is almost beyond empirical doubt.9

The interaction between epistocracy and structural inequality generates three mutually reinforcing objections to the idea that we ought to ‘let the wise rule’: ‘Homogeneity’, ‘Perspectives’, and ‘Practices’. Together, they undermine EA (1) and the epistocratic project.

  • . Homogeneity

Epistocracy requires the ability to delineate between the wise and the unwise in an epistemically valuable way. Otherwise, it cannot harness the ‘Value of Wisdom’ or mitigate the ‘Burden of Ignorance’.

Existent structural inequality makes such an ability unlikely to obtain, by ensuring epistemically-damaging homogeneity amongst an epistocratic group.

Typically, epistocrats propose to distinguish the wise from the unwise via proxy or a sufficiently difficult political knowledge test. 10 The problem is that political knowledge or access to suitable proxies (e.g. education) is asymmetrically distributed in favour of the materially and/or socially advantaged. 11 This makes sense in the context of structural inequality: sexist and racist education practices (both formal and informal), amongst other structures, have ensured that political knowledge or access to it is the preserve of the most advantaged.

The upshot of this is clear: the epistocratic, ‘wise’ group will be fairly homogenous. After all, the materially and socially advantaged are more likely than others to pass the ‘test’ or satisfy any suitable proxies employed.

Exactly why homogeneity is epistemically damaging will become clear in Sections IV and V. For now, it is sufficient to note that homogeneity is something an epistocrat wants to avoid if the ‘wise’ group is to be epistemically valuable.

The epistocrat might respond in two ways. Most obviously, they might claim that any suitable political knowledge test or proxy would capture the political wisdom of the disadvantaged. Perhaps a test created by a sufficiently large, randomly selected group might be the best way to do this.

This solution merely pushes the problem up a level. Inequality would influence the creation of the epistocratic group indirectly since the materially advantaged could pay for coaching, or familiarisation with test material etc. The problem of homogeneity would persist.

The second response might be to temporarily bite the bullet. Although currently homogenous, an epistocratic group would become diverse over time because epistocracy would effectively work to address structural inequality.

The main problem with this response is naïvety. The next two objections suggest why: epistocracy, because its essential maintenance of political structural inequality, is unlikely to effectively address issues important to the most disadvantaged. Even setting this aside, however, the response is costly. After all, it implicitly concedes that creating an epistemically valuable epistocratic group is implausible in a structurally unequal society.

  1. Perspectives

A deeper problem with epistocracy is that it is fundamentally mistaken about the extension of the class of ‘epistemically valuable perspectives’ (EVPs) in decision-making; this is especially true in a structurally unequal society. Epistocracy essentially limits the class of EVPs to those of the epistocratic group and systematically silences the perspectives of the non-epistocratic group via relative or absolute disenfranchisement,.

All perspectives are EVPs in decision-making because of the asymmetric distribution of political outcomes: all decisions produce outcomes, each of which is disproportionately experienced by some groups in society but not others. We can distinguish between two kinds of asymmetric distribution:

Weak Asymmetric Distribution (WAD): Everybody in society directly experiences at least one outcome of the decision, but nobody directly experiences all of the outcomes.

Strong Asymmetric Distribution (SAD): Not everybody in society directly experiences at least one outcome of the decision.

Recognising this suggests that all perspectives are always at least somewhat valuable, since different groups experience different outcomes of the same decision. Consequently, different groups have greater epistemological access to the values of certain outcomes than others. For example, the migrant family will experience different outcomes from the election of a fiscally-conservative, socially-conservative new ruling party than the wealthy, native banker will. In turn, the migrant family are better placed to understand the value of migration-related policies than the banker is.

Including both perspectives in this case is epistemically valuable in two related ways. First, our cumulative epistemological access to the total value of outcomes produced by the decision increases, since no group has identical access. Second, the potential for ‘epistemic blindspots’ is reduced: by including all perspectives in decision-making, we will not incidentally exclude an especially valuable perspective with respect to the value of a decision’s outcomes.

These considerations gain force when we think about decisions with a SAD of outcomes. After all, not all groups experience any outcome of the decision. Therefore, the perspectives of those who do are especially epistemically valuable in understanding outcomes’ values.

A similar point applies when we consider structural inequality. Because of inequality, the outcomes of the same decision are often experienced in radically different ways. For example, the materially advantaged experience the election of a fiscally-conservative party in a completely different way to the materially disadvantaged. In a more equal society, experiences might not diverge so radically.

Assuming justifiably that experience increases epistemological access at least in value-questions, it is true that all perspectives are EVPs in decision-making, since outcomes (and with it epistemological access) are asymmetrically distributed. This is a problem for EA (1): if epistocracy cannot harness all EVPs because it disenfranchises some, it is unlikely that it can accurately appreciate the value of outcomes. In turn, this makes it less likely to routinely produce valuable outcomes.

The epistocrat might attempt three responses. Least successfully, they might attempt to deny the ascription of epistemic value to experience. This is unlikely to persuade many: the value ascription is minimal and, anyway, such an ascription is commonly accepted in value questions.12

Second, they might accept the premise, but claim that an epistocratic group can contain all EVPs. Again, this response is unpersuasive. For one thing, the ‘Homogeneity’ objection suggests that this is implausible in our non-ideal society. More importantly, it is conceptually impossible that epistocracy represents all EVPs. Non-epistocratic perspectives are also EVPs, especially with respect to decisions that only impact them. For example, a policy which made night school compulsory for the non-epistocratic group would only directly impact the non-epistocratic group. On the ‘Perspectives’ argument, their perspectives would certainly be EVPs. However, epistocracy definitionally excludes their perspectives in decision-making.

The third response to the ‘Perspectives’ objection, then, is to deny the underlying assumption that formal disenfranchisement entails a loss of perspective. The ‘Practices’ objection undermines this response.

  1. Practices

There are two good reasons to think that formal disenfranchisement will lead to loss of EVPs, both relating to the ‘labelling’ essential to epistocratic disenfranchisement.

Labels structure responses. If something is publicly labelled ‘bad’, respondents are more likely to disregard it. Similarly, when epistocrats publicly label some ‘unwise’, they invite disregard for their views in deliberation. Their epistemic credibility, ability to solve complex problems or, generally, their perspectives, are all likely to be devalued.

The epistocratic might deny that a ‘wise’ group would engage in devaluation. However, devaluation is subconscious and difficult to mitigate. It is comparable to the ‘marketing placebo effect’ in that epistocrats might not consciously change their response based on labels, yet it occurs nonetheless. 13

Devaluation also occurs amongst the non-epistocratic group as a result of labelling. The stigmatising effect of being labelled ‘unwise’ by the state is likely to result in a loss of epistemic self-belief. As Haslanger notes, it is difficult to retain even justified beliefs in the face of devaluation.14

Of course, the epistocrat might dismiss this as a semantic issue, and change their labels accordingly. However, this does not address the root of the problem. The fundamental division between the two groups is what fuels epistemically damaging perspectives.

  1. Conclusions

We ought not let the wise rule, at least not in a structurally unequal society. There is little reason to believe EA (1): against a background of structural inequality, an epistocratic group is likely to be homogenous and unable to capture epistemically valuable perspectives. In turn, this makes it unlikely that it will ‘tend to produce better outcomes’ than democracy.

For now, we ought to retain faith in the ‘mob’, at least in the absence of a credible alternative.

 

Endnotes:

1 See e.g. Plato in Republic Book VI

2 e.g. Brennan, J. (2016), Brexit, Democracy, Epistocracy; see also https://foreignpolicy.com/2016/11/10/the-dance-of-the-dunces-trump-clinton-election-republican-democrat/

3 e.g. a wisdom-based plural voting scheme.

4 See Landemore (2017) for a discussion of Rawls’ (1993) objection to objectivism and/or cognitivism about some political claims, Habermas’ (1990) response and Martí’s (2006) argument.

5 e.g. Landemore, Cohen, and Estlund (democrats); Ingham (2013) and Brennan (2011), (2016) (epistocrats).

6 e.g. Estlund (2008)

7 https://www.newyorker.com/magazine/2016/11/07/the-case-against-democracy (Accessed:22/01/2022); https:// www.ipsos.com/en-uk/perceptions-are-not-reality (Accessed: 22/01/2022)

8 This list is not exhaustive.

9 See e.g. the gender pay gap; access to better education dependent on material resources; attainment gaps along class lines, etc.

10 E.g. Mill (1861) or Brennan (2016) and https://www.vox.com/2018/7/23/17581394/against-democracy-book-epistocracy-jason-brennan (Accessed: 23/01/2022)

11 See Bhatia (2020) for a discussion of relevant empirical studies supporting this.

12 e.g. questions about artistic value.

13Schmidt et al (2017)

14 Haslanger (2014)

  • Facebook
  • Twitter
  • Reddit

6 Responses to Oxford Uehiro Prize in Practical Ethics: Why Don’t We Just Let The Wise Rule?!

  • Paul D. VanPelt says:

    Why NOT let the wise rule? The various arguments are appealing, for various reasons. A sceptical pragmatist sees the utility of some over that of others. Other thinkers are perhaps wiser when recognizing one ancient impediment.: status quo. Near-classical judgment on this aspect effectively says: we had better keep what we are mutually uncomfortable with. The outcome from tinkering might be worse. Looking at parts of this world and governance thereof, produces clear and convincing evidence. Better to do what has always been done and getting what has always been gotten, than to chance something different. Current modes of democracy are state-of-quo predictable (I thank my wiser brother for that phrase, state-of-quo). Chief executive officers, and downward on the chain-of-command, will act within the spirit, if not the letter, of law. Well, most of them, radar blips notwithstanding. Aside from certain fabrications, citizens know fairly well what they are getting—(see previous sentence). Our legislators enter the ring to make money. Saying that, it must be fairly noted they are not against our doing so as well. More can be said—already has been.

    • L says:

      Ratio

      • Paul D. Van Pelt says:

        Don’t know what you mean with the one word comment. Too cryptic for my limited experience. If, and only if, you are using it to illustrate inertia, then one might infer your remark to support the status quo rationale. Conservatively speaking, it is better to retain an inadequate stance than adopt a different one which could make things worse. Case-in-point: today’s confirmation of associate justice Jackson. After the vote, it was clear to anyone paying attention that Mr. Paul had not voted—was not present when his vote was called. The outcome was decided by that time, but the process was suspended, in deference to the legislator and his right to lodge his vote…in effect, giving him benefit of doubt; tolerating his impedance of established procedure. Barring a health emergency or other unavoidable circumstance, his tardiness could be viewed as partisan obstruction, even in light of the already-decided outcome. Petulance, unbecoming a government official is not unheard of. Censure is also regrettable, but not unknown. Paul is unlikely to be censured for his conduct. But, all it would do would be to make things worse.

  • Pavel Novak says:

    All the theories that try to “invent” the ideal political system has wrong base or let us rather say wrong assumption.

    They think that the politics has the clue.
    But it does not.
    The politicians are not supermen. And all the governments fail.

    No political regime is able to solve the problem. Even the authoritarian or totalitarian regimes that gained absolute power were helpless
    to solve the problems that the society had.

    So the purpose of the political system is not to ensure that the best decisions will be made or the best people will rule.
    It would be just the kind of Plato’s uthopia. We simply cannot force the best people to go to the politics. It is not freedom.

    So as already Huntington says: THE PURPOSE OF DEMOCRACY IS NOT TO ENSURE GOOD RULE. THE PURPOSE OF DEMOCRACY IS TO ELIMINATE TYRANNY.
    No less, no more.
    The aim of democracy is only to change the politicians in regular terms, not to solve problems.
    Yes…. You can say it is not too much.
    But in big countries like Russia or China people do not have this option….
    So do not try to change democracy. The only result of such effort will be some kind of dictatorship.

    • Ian says:

      Truths end? Whilst democracies do apparently degenerate into tyrannies as part of a form of a growth/decline process I disagree that the purpose of democracy is only to prevent tyranny. Perhaps one of the objectives in maintaining democratic process is, but not the whole purpose. The purpose of democracy more frequently appears as a means of maintaining the broader truth of the social life. Tyrannies on the other hand require only a narrow truth which becomes enforced in various ways, frequently increasingly desperately. So the answer to the question posed becomes of paramount importance in determining where any given society is going for the quantity of individuals whose wishes they are either responding to, or imposing a particular outcome upon. For any member the quality of the social life would probably be influenced by that mix.

  • Pavel Novak says:

    By eliminating tyranny the democracy ensures that members of society, i. e. INDIVIDUALS, not society itself, has broad choice to develop their life projects.
    Democracy is inherently connected with less range of violence and on the contrary with more range of freedom.
    By this means democracy ensures “the truth of the social life”. Therefore no added value to democracy is necessary.

Leave a Reply

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

You may use the <em>, <strong> and <blockquote> tags. Links have been disabled to combat spam.

Notify me of followup comments via e-mail. You can also subscribe without commenting.

Authors

Affiliations