Two epistemic accounts of democratic legitimacy

Polity

Volumn 37, Issue 2, 2005, Pages 216-234

  Hershenov, David B ^a  

^a UNIVERSITY AT BUFFALO   (United States)

Author keywords

Democratic legitimacy; Epistemic democracy

Indexed keywords

EID: 19944392899     PISSN: 00323497     EISSN: None     Source Type: Journal    
DOI: 10.1057/palgrave.polity.2300012     Document Type: Article

References (44)

1. What is meant by a "fair procedure" could be a number of things. In the case of a fair vote, what people have in mind is usually something along the lines of everyone gets one vote, the ballot is secret, votes aren't bought, only residents vote, and, very important to advocates of deliberative democracy, citizens are able to participate in pre-election deliberations and do so with enough resources in order that their view gets a reasonable hearing and a good faith debate ensues between them and their opponents. On some accounts, fairness will also involve restrictions on the type of reasons (public reasons a la Rawls) that can be introduced for consideration.
2. I am using "reasonable" and "rational" interchangeably throughout this paper. Rawls, famously, distinguishes the two.
3. The position that the majority may be more likely to be correct is of course not new. Condorcet, Bentham and Rousseau all held something along this line. For an account of Condorcet's Jury Theorem see footnote 6. While Bentham distinguished correct answers (in accordance with the Principle of Utility) from majority opinions, he did claim that "general consent provides the surest visible sign and immediate evidence of general utility" Cited in Ross Harrison, Bentham (London: Routledge and Kegan Paul, 1983), 214. And according to Joshua Cohen in his "Epistemic Populism,"
4. Ethics (October 1986): 28, "While Rousseau thought that the general will - which aims at the common good - 'is always right,' he denied that it 'follows that the people's deliberations always have the same rectitude' since those deliberations may reflect insufficient information, or be dominated by private interests, or subordinated to factional conflict, or addressed to issues on which there is not common interest. He concluded that majority judgments are good if fallible indicators of the general will under certain specific background conditions, including good information, widespread political participation, absence of factions (or else a multiplicity of factions), limited economic inequality, and the rule of law And even with these qualifications, Rousseau did not endorse rule by simple majorities. Rather, he thought that, on more fundamental issues, it would be reasonable to require larger majorities. In fact, only on 'business matters,' as distinct from laws, is it in general true that simple majorities should be decisive."
5. See David Estlund's critique of Rousseau's demand that those in the minority relinquish their epistemic autonomy and concede that we were wrong in his "Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority" reprinted in Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg (Cambridge, MA: MIT Press, 1997), 198-200.
6. And this agnosticism extends to moral realism. Or more accurately, some of the reasons for the latter agnosticism motivate the agnosticism regarding the epistemic accounts of democracy
7. Condorcet's Jury Theorem, on the other hand, maintains that a mere 51% of the vote is good reason to believe a view is true if people are right more than half the time. Estlund offers a nice account linked to the "Law of Large Numbers" of what Condorcet's theorem implies. "If each voter has an individual likelihood above 50% (call it [50 + n] %) of giving the correct answer (whatever it is) to a dichotomous choice (heads/tails, yes/no, true/false, better/worse, etc.), then in a large group the percentage giving the correct answer is bound to be exceedingly close to (50 + n)%. Therefore, the chance that it will be at least 50% is even higher, approximating certainty as the group gets larger or the voters are better. In summary, if voters are all 51% likely to be correct, then with a large number of voters it is almost certain that almost exactly 51% will be correct, and so even more certain that more than 50% will be correct." "Beyond Fairness and Deliberation," in Deliberative Democracy, 202-03. Because Condorcet's theorem works with a mere majority, some readers might find my first eptistemic account less appealing. But the Jury Theorem has its well-known problems. Its axioms may not hold in a complex modern democracy.
8. See Estlund,"Beyond Fairness and Deliberation," 186-91
9. and T Christiano's "Freedom, Consensus and Equality in Collective Decision Making," Ethics (October 1990): 151-81.
10. For fuller accounts of deliberative democracy see the volume Deliberative Democracy, especially Joel Cohen's contribution "Deliberation and Democratic Legitimacy," 67, 72-73.
11. See Estlund's "Beyond Fairness and Deliberation" for a discussion of procedure and procedureindependent epistemic approaches to democracy, 176-81.
12. Cohen. "Epistemic Populism," 34.
13. For a confusing array of sample usages of "reasonable" and "rational" which may leave one believing if the terms are not ambiguous they are then what Wittgenstein called "family resemblance" terms, see the various offerings in the Deliberative Democracy anthology.
14. For skepticism about overcoming the plurality of moral views that characterizes modem societies through democratic deliberations, see Gerald Gaus's "Reason, Justification and Consensus: Why Democracy Can't Have it All," 207, 215;
15. and Christiano's "The Significance of Public Deliberation," 250, 264, 266-70. Both are in Delibemtiue Democracy.
16. See also Christoper McMahon's review of Deliberative Democracy in Ethics (April 1999): 648-50.
17. If one thinks that there can't be unknowable truths, for they imply concepts we can't grasp, which Davidson showed was incoherent in his "The Idea of a Conceptual Scheme," I suggest they read Fodor's and Nagel's responses, respectively, in The Modularity of Mind (Cambridge, MA: MIT Press,1983), 120-24
18. and The View From Nowhere (Oxford and New York: Oxford University Press, 1986), 93-98.
19. Crispin Wright, Truth and Objectivity (Cambridge, MA: Harvard University Press, 1992).
20. One was that there are moral truths. A second big "if" was that my argument hangs on there being a justifiable conception of practical rationality which others might doubt.
21. Although "inconclusive" is used above in the sense that we didn't achieve unanimity, the deliberative process could have been constructive in revealing flaws in one side's reasons, which thus makes the other view more attractive by default, or perhaps improved both sides but neither to the point where it appears to persuade the other side, or revealed in a better light the opponent's principles. All of these could make what originally looked like a disagreement where one side maintained that the other was irrational into a disagreement where both sides view the other as rational or, at least, an impartial outsider can view them as "rationally and morally tied."
22. Again, this is supposing the involved parties believe in morality that can be intuited. If not, then the following will not be persuasive.
23. But this may in part be due to the fact that the others previously agreed with oneself.
24. J.L. Mackie advocated such abandonment in a section entitled "The Argument from Relativism" in his Inventing Right and Wrong,
25. reprinted in Essays on Moral Realism, ed. Geoffery Sayre-McCord (Ithaca, NY: Cornell University Press, 1988), 109-10.
26. I believe that the intuitionist view which I am putting forth would be compatible with Alvin Plantinga's account of non-inferential basic beliefs that have warrant and thus count as knowledge when they are produced by faculties functioning properly in the environment that they were designed for. See his account in Warrant and Proper Function (Oxford: Oxford University Press, 1993).
27. Perhaps it would be better to say "moral blind-spot" than "moral blindness," for the latter appears to suggest more blindness than the former. But maybe someone completely unethical or amoral should be considered morally color-blind.
28. As Thomson puts it, if any moral claim is true, it is that torturing babies for fun is wrong; Judith Jarvis Thomson, Introduction to her The Realm of Rights (Cambridge, MA: Harvard University Press, 1990), 18.
29. A related criticism pointed out by C. McMahon is that if the majority is under the sway of a single charismatic leader, then there is little reason for the minority to heed the will of the larger group for the former is in a sense in a disagreement with but a single person. It is as if one person has been cloned enough times to form a majority So my argument is the strongest where a diverse group of people independently reaches the same decision. But McMahon, in a written communication, responds that this leads to a kind of paradox. He writes, "If the views of those on the other side have been arrived at independently, then they have paid little or no attention to what others think is reaching them. They have made up their own minds on the basis of the evidence. There is a kind of paradox here. The less attention others have paid to what other people believe, the stronger the reason provided to me by the fact that many other people hold a particular position. But if this is the case, it would be odd for me to respond to this reason by simply assuming I am wrong. The response that is most in keeping with the way the reason arises would be to reexamine the issue to see if I can discover any mistakes I have made. But if I cannot, I should hold on to my view. I should not conclude that others are probably right just because so many of them hold the same view. This would be to reject in my own case the independence that is the sin qua non of the agreement of many others having any rational significance at all."
30. The many different models of calculators is comparable to the diversity of the moral community in the previous passage.
31. See his article "Values and Secondary Qualities" reprinted in Essays on Moral Realism, 166-80.
32. Again, just as chances are that the minority is more likely than the majority to perceptually deviate and thus make a mistake in color perception, so too with the moral analogy. Note, however, that this probabilistic claim doesn't contradict what I said earlier about it not being more probable that a rational group of inquirers would attain the truth. The antiprobabalistic claim there was due to my doubts that there were any moral truths. The probabalistic aspects of my epistemic account here rests on the assumption that there are moral truths.
33. Gilbert Harman takes this approach as does John Mackie. Both are reprinted in Sayre-McCord, Essays on Moral Realism. Harman's account is on pages 119-26.
34. This is a version of a criticism that an anonymous reviewer put forth.
35. According to Estlund, a coin toss may be the paradigm of fairness. See his "Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority" reprinted in Deliberative Democracy: Essays on Reason and Politic, 176-79.
36. See, for example, Shelby Steele's The Content of our Character (New York: St. Martin's Press, 1990), 111-26.
37. This entrenched minority power play (of academics, administrators and students) is the equivalent of whites being opposed to affirmative action merely because it takes away opportunities from them.
38. This might not be the case for those employed in certain jobs where intelligence above a certain level normally has little impact on job performance, since the work is not intellectually demanding.
39. Dworkin makes this point in his "Reverse Discrimination" in his Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1977), 223-48, and in his writings on the Bakke case
40. reprinted in A Matter of Principle (Cambridge, MA: Harvard University Press, 1985), 293-315.
41. Admittedly, there are some contestable counterfactual assumptions being made here.
42. Chris McMahon has expressed to me his skepticism about my account of a "rational tie." He doubts that it accords with the phenomenology of moral disagreement. In his view, it is unlikely that a prolonged tie will lead to the conclusion that there isn't a sought-after unique right answer. The opposing parties are more likely to insist that the other is wrong, even if they can't presently point out the cognitive failure the other side suffers from. An anonymous reviewer made a similar point.
43. Well, one problem is that the majority could still get their way by voting for the position they don't really want since they know the minority will triumph when there isn't a full consensus. But this can be avoided if people are really voting for the public good, which we stipulated they would do for the sake of discussion or by randomly altering the majority/minority "sweepstakes."
44. This is likely to be the case with affirmative action. Another example would be if members of a poor minority ethnic group were the ones regularly being sentenced to death rather than life imprisonment.