Deliberative Democracy: A “Neutral” Result is a Negative Result

Chapter 3 of Against Democracy looks at the empirical work on how political participation, and in particular, deliberative democracy, affects us. Mill hypothesized that deliberation would tend to educate and ennoble us. But, it seems, it tends to corrupt and stultify us.

In the end, I argue the empirical evidence is much more damning that people realize. This excerpt from a draft explains why:

As we saw, there is ample empirical evidence that deliberation often stultifies or corrupts us, that it often exacerbates our biases and leads to greater conflict. However, for the sake of argument, suppose none of this evidence existed. Suppose, instead, that all we had were neutral results. That is, suppose empirical political scientists had continually tried to test the thesis that deliberation educates and enlightens, but continually failed to find evidence that it does. In that case, it would be tempting to conclude that deliberation is pointless and ineffective, but at least not harmful.

Researchers often present their findings this way. Sometimes, researchers say that while they didn’t find positive results, they at least didn’t measure negative results. The results were neutral.

On the contrary, I’ll argue here, a neutral result is usually a negative result. If people deliberate together, but this fails to educate or enlighten them, then this means they are actually worse off as a result of deliberation. If I am right, then the empirical work on deliberative democracy is much more damning than other philosophers, political theorists, and political scientists have realized.

What is rational for you to believe or not to believe depends upon the evidence available to you. Imagine a child has led a sheltered life, with no exposure to history, geology, biology, physics, or cosmology. She believes, on the basis of her young Earth creationist parents’ testimony, that the universe is 6,000 years old and that all animals were created 6,000 years ago. But suppose this child then takes sixteen years of classes in history, geology, biology, physics, and cosmology. Along the way, she gets to sequence DNA, re-create Mendel’s pea experiment, handle fossils, and the like. Yet, after sixteen years of intense study, suppose she continues to believe the world is 6,000 years old and that all animals were created as they currently are.

In this case, from an epistemological standpoint, she got worse. After all, she encountered an overwhelming amount of evidence confirming evolution and disconfirming young Earth creationism. She should have changed her mind, but didn’t. After sixteen years of study, the gap between A) what she believes and B) what she ought to believe increased. Her beliefs are less justified now than they were sixteen years ago, before taking the classes and doing the experiments. She has thus violated her epistemic duties. She added further wrongdoing to her epistemic tally sheet. She is more epistemically delinquent after getting new evidence than she was before. In that case, it would be a mistake to report that taking classes had a neutral effect on her epistemic situation. She’s actually worse off.

Now consider what happens during deliberation: When someone learns that her other smart, well-informed people disagree with her about some issue, she might question whether she should reduce her confidence in her own beliefs.[i] If she encounters new information and evidence, she should revise her beliefs accordingly. Even in badly run, badly functioning deliberations, most citizens encounter new arguments and new information, arguments and information that should cause them to revise their beliefs or weaken their degree of confidence. Citizens should weigh other citizens’ testimony on the basis of how expert, reasonable, and reliable those citizens are likely to be, and revise their own beliefs accordingly. If the citizens do not revise their beliefs accordingly, then their epistemic situation has worsened. Deliberation made them more delinquent.

Thus, when deliberation has no effect on citizens’ beliefs or their degree of credence in their beliefs, we should generally interpret this as showing that deliberation made them worse, from an epistemically point of view. Just as a university-educated Young Earth Creationist is epistemically inferior to an uneducated Young Earth creationist, so a person who does not revise his beliefs or degrees of belief after deliberation is (usually) epistemologically inferior to his situation before deliberation.

Deliberative democrats must conclude that similar remarks apply to citizens’ moral status post-deliberation. Deliberative democrats usually hold that the rules of proper deliberation are moral rules. They believe citizens have moral duties to abide by the rules of deliberative democracy. In their view, citizens are obligated to deliberate properly. Thus, if we find that most citizens are not deliberating properly, the deliberative democrat should conclude that the gap between what the citizen A) ought to have done and B) did in fact do has widened. The citizen has added further moral wrongdoings to her lifetime moral tally sheet. After deliberation, she is defective from a moral point of view than she was before.

[i] Feldman 2006; Elga 2007.

Published on:
Author: Jason Brennan
  • Rob Gressis

    Couldn’t it be that co-deliberators’ conclusions don’t change, but that either the reasons on which they base their conclusions, or the confidence with which they hold their conclusions, do? In that case, they could be becoming epistemically more virtuous even though they still hold the same beliefs.

    • Jason Brennan

      Here we’re imagining that this doesn’t happen either. The point is to say tat if deliberation had no effect, we should view that as a negative result.

      But empirically it does have lots of effects, and most of the actual effects are negative. E.g., empirically, what tends to happen is that when people who believe X get evidence that not-X, they actually strengthen their confidence in X.

  • Your argument here is vitiated by the falsity of your claim that

    “What is rational for you to believe or not to believe depends upon the evidence available to you.”

    I criticise that claim in a short article available here:

    and in more detail in these papers:

    In short, you have a misconception of rationality. Being rational does not mean being a passive, unbiased receptor of evidence; it means being active in questioning available theories, including ones own, and also in questioning evidence (all evidence is theory-laden).

    I am no fan of deliberative democracy. Being a libertarian of sorts, I think that it is illegitimate because it involves disputes over the particular ways in which people’s rights are to be violated. I agree with you that it often leads to internecine strife rather than mutual illumination. However, I think that that failing of it is a contingent one. If people engage in discussion with a view to learning from each other, and do so in a rational, i.e. questioning, way, then they can usually learn something new. That does not mean that they end up agreeing; it means that they end up with better positions than they had previously, even if they still disagree.

    One of the big problems in contemporary philosophy is justificationism. The idea that we should or could have justified beliefs belongs to a primitive, pre-scientific mentality. It leads to the ideas, so often mouthed by contemporary philosophers, especially those who advocate so-called ‘critical thinking’ (it is anything but), that the point of an argument is to establish a proposition, and that the aim of discussion is to persuade. We get a better understanding of argument if we see its point as being to show logical inconsistencies, and of rational discussion if we see its aim as to learn from each other. I have a paper on this too, here:

    also section 8 of my paper on ethical intuitionism, which criticises Huemer’s ‘phenomenal conservatism’:

    • Jason Brennan

      “What is rational for you to believe or not to believe depends upon the evidence available to you.”

      I would have thought this was the least controversial thing one could possibly say about epistemology, but, no surprise, in philosophy, there’s a challenge to everything. Will take a look at your papers.

      • CJColucci

        I’m not a professional, or even competent amateur, philosopher, but I’ve known for decades that this is far from uncontroversial.

      • Goose

        Maybe if you substitute “rational” with “logical”? “consistent”?

        Using DF’s framework, “rational” as an adjective for a person’s belief(s) goes out the window – it’s more of a personality trait/approach towards inquiry.

  • Jameson Graber

    I think you make a good point, here, but I’m really turned off by the idea of morality as a tally sheet.

  • Chris Thomas

    There’s an additional reason to interpret neutral results as negative if we view deliberation as a cost of producing knowledge. If we deliberate for hours, or years as in Jason’s example above, and nothing about our knowledge changes, then we have incurred a large cost with no benefit, clearly a bad thing. Consider a paper looking at the effects of working on income. The paper finds that on average, working has no effect on income. We would obviously conclude that working is a bad thing to do, since again, you incur real costs, but get no benefit, a net loss. Of course, this argument does not work if you consider deliberation a consumption good.

  • Pingback: Deliberation: The Double Edged Sword | The Indian Economist()