Bad Beliefs: Why They Happen to Good People

Click to increase image sizeClick to decrease image size

Notes

1 Such pollution is the effect of what one may call ‘epistemic weapons’. See Pettigrew (2022).

2 I thank Richard Pettigrew for discussion of this point.

3 All page-numbers in parentheses without further reference are to Levy's book.

4 I avoid ‘good and correct belief’ because of the moral and ethical associations these terms would carry.

5 He does not consider that these belief battles may be proxies for battles over diverging interest. I return to this below.

6 In his critical discussion of rationality deficit accounts Levy does not directly challenge the (rather thin) view of rationality these accounts presuppose. This critique is, for example, an important feature of the strategy of Rizzo and Whitman (2019), especially chapters 2–3. Indirectly Levy often seems to rely on an ecological view of rationality (130–131; 142–143; 149–159), but this is not necessary to his argument. He has informed me that he thinks that ‘a process is rational in virtue of how it processes information’, and though he never uses the word in the book, he is ‘actually a Bayesian of some sort, and Bayesianism is a paradigm non-ecological view of rationality’. (Communication to the author, July 29, 2022)

7 This meets the behavioral economist, who uses deviations from a rational choice model as her strategy to make claims about purported irrationality, on her own turf.

8 We can find an anticipation of something close to this idea in Rizzo and Whitman (2019, 418): ‘we find it peculiar (at best) to treat mere provision of information as necessarily paternalistic … from the perspective of inclusive rationality, the focuses would be on the accuracy of messages – both in content and interpretation’. (emphasis in original) It's the very importance of social cues that makes them attractive to would be manipulation.

9 A century ago Russell had noted the same possibility within an imported, scientific culture:

the manner of life produced by science can be taken over by populations which have only certain practical rudiments of scientific knowledge; such populations can make and utilize machines invented elsewhere, and can even make minor improvements in them. If the collective intelligence of mankind were to degenerate, the kind of technique and daily life which science has produced would nevertheless survive, in all probability, for many generations. But it would not survive forever, because, if seriously disturbed by a cataclysm, it could not be reconstructed. (Russell 1928, p. 35.)

10 I am thinking of, say, cultural schismogenesis as discussed by Graeber and Wengrow (2021).

11 In wider context, Levy cites Millgram’s (2015). This book also draws on Oreskes & Conway, as well as Latour.

12 Levy's examples are drawn from recent politics, but he could have mentioned fashion and shifts in market sentiments in financial markets.

13 In the case of modern financial markets these beliefs are also outsourced to algorithmic devices.

14 Somewhat surprising Levy does not use Clark's ‘predictive brain’ hypothesis to strengthen his argument. This, too, tends to treat internal representation as thin. See Clark (2013).

15 Zaller (1992). See also Downs (1957). Downs would have been familiar with Knight's footnote that ‘It is evident that the rational thing to do is to be irrational, where deliberation and estimation cost more than they are worth’ (Frank Knight Risk, Uncertainty, and Profit. Hart, Schaffner, and Marx; Boston Houghton Mifflin, 1921, p. 67).

16 Downs (1957), op. cit. pp. 139–140.

17 I thank Nathan Ballantyne for putting it like this.

18 Republic, 488a-d. Arguably while defending the populace's capacity to choose its leaders, Machiavelli makes the same point in his attack on the grandi in Discourses on Livy. See Bellamy (2018).

19 As Richard Pettigrew noted to me, ignoring outlying data ‘seems less to do with expert deference and more to do with good statistical reasoning. The paradigmatic argument for this point “is Hume on miracles!’” In addition on my view, the most robust sciences are capable of turning outlying data, even discrepancies, into higher quality evidence or second order evidence. For striking examples, see Smith (2014).

20 Stigler (1975, pp. 3–4). Unlike Stigler and Levy, think keeping intellectual outliers within a scientific discipline can be defended on epistemic and moral grounds: Lefevere and Schliesser (2014).

21 I return to this below. For a recent survey that explores such issues in the vernacular of formal philosophy, see Dorst (2020). Dorst also notes that there are different kinds of higher order evidence. I thank Richard Pettigrew for calling my attention to it. See also Lasonen-Aarnio (2014) for the existence of important constraints on all strategies to deal with dilemmas generated by higher order evidence.

22 Levy personal correspondence with the author, 29 July 2022.

23 I thank Maarten Boudry for the suggestion.

24 This is already thoroughly documented in Carson (1962). Arguably the significance of Leviathan and the Air-Pump was its argument that science was always entangled with power and politics.

25 As I was drafting this review Science (377: 6604), 22 July 2022), published a paper, ‘Blots on a Field’, by Charles Piller that shows how ‘A neuroscience image sleuth finds signs of fabrication in scores of Alzheimer's articles, threatening a reigning theory of the disease’. https://www.science.org/doi/epdf/10.1126/science.add9993. See also, Marek, Tervo-Clemmens, Calabro et al. (2022).

26 Schliesser. ‘Galilean Reflections on Milton Friedman's “Methodology of Positive Economics,” with Thoughts on Smith’s (2005). For more sophisticated versions of this point, see Stegenga (2018). And for a polemical example, see Lemoine (4 December, 2020).

27 I owe the insight and the formulation to Jeffrey Friedman, personal correspondence 15 August, 2022.

28 Kuhn (1970). Aumann (1976). On the pre-history of this idea within philosophy and political economy, see Schliesser (2015).

29 https://www.ipcc.ch/about/structure/, Accessed 28 July 2022.

30 In context, Levy's own point may be orthogonal to the use I make of the quote.

31 See Rizzo & Whitman, op. cit.

32 Of course, it does leave open how we should think about the effectiveness of nudging. See Maier, et al. (2022).

33 Personal communication with the author, 29 July 2022.

34 I thank Boudry for discussion.

35 Personal communication with the author, 29 July 2022.

36 Levy generously cites my Schliesser (2019). I have not claimed originality for this approach to philosophy (or its name which goes back to Spencer).

37 Levy is silent on how, exactly, to distinguish analytic from his synthetic philosophy, but it is clear that he is not interested in careful conceptual clarification with detailed cautious arguments, but rather drawing on the sciences to create a theory that can improve the world.

38 Levy cites Ballantyne (2019).

39 The Locus Classicus is Peart and Levy (2009).

40 Crucially, the expression of genetic differences are on this view also triggered by the environment.

41 Levy ignores that within science, even climate science, this dependency on others is not uncritical. As Michael Polanyi urged, scientists from nearby disciplines and related skillsets do act as an epistemic check on uncritical acceptance. See Schliesser and Winsberg (2020) and Winsberg (2018).

42 David M. Levy, ‘Gordon Tullock and the Predatory Economist’ in Matters, Liberty. ‘Peter Boettke, Gordon Tullock and the Rational Choice Commitment (November 2017)’. https://oll.libertyfund.org/page/liberty-matters-peter-boettke-gordon-tullock-rational-choice

43 See, especially, Goodin and Spiekermann (2018). I thank Richard Pettigrew for raising the question with me.

44 I am not suggesting he is the first in the European philosophical tradition to make it central to his theory. In Book 2 of Plato's Republic 372a-c, the division of labor is key to the origin of political life as described in the so-called true city or city of pigs. The Kallipolis clearly also has a cognitive division of labor.

45 I note here the gendered nature of Smith's account of philosophy. (Levy is also silent on the gendered division of cognitive labor.) Also, sometimes Smith uses ‘philosophy’ in a more capacious sense (which includes all sciences and learning), but here ‘philosophy’ is clearly distinct from some of the more practical sciences.

46 See especially his treatment of casuistry as expertise-for-hire. Schliesser (2013).

47 For more on the role of government (and while drawing on Lippmann's views) to help combat epistemic pollution while sticking to liberal principles, see Nick Cowen & Eric Schliesser (ms) ‘The Articulate State’.

48 I thank Erwin Dekker for pressing this point on me.

49 I am grateful to Ryan Muldoon for encouragement. I thank Nathan Ballantyne, Richard Pettigrew, Adrian Bardon, Jeffrey Friedman, Maarten Boudry, and Helen de Cruz for detailed comments on an earlier draft. I also thank Neil Levy for clarifying a number of features of his view in personal correspondence with me. The usual caveats apply.