Abstract
Modern societies are characterized by a division of epistemic labor between laypeople and epistemic authorities. Authorities are often far more competent than laypeople and can thus, ideally, inform their beliefs. But how should laypeople rationally respond to an authority’s beliefs if they already have beliefs and reasons of their own concerning some subject matter? According to the standard view, the beliefs of epistemic authorities are just further, albeit weighty, pieces of evidence. In contrast, the Preemption View claims that, when one discovers what an authority believes, it is not permissible to rely on any of one’s own reasons concerning the subject matter. The original version of this view, as proposed by Linda Zagzebski, has recently been severely criticized for recommending blind trust and for abandoning even minimal standards for critical thinking. In our paper, we defend a new version of the Preemption View—Defeatist Preemptionism—in a way that differs radically from Zagzebski’s. We argue that our view can be derived from certain widely accepted general epistemic principles. In particular, we claim that preemption can be identified as a special case of source sensitive defeat. Moreover, we argue that Defeatist Preemptionism does not lead to the undesirable consequences that critics ascribe to the Preemption View. The paper thus articulates the foundations and refinements of the Preemption View, such that it adequately captures the phenomenon of epistemic authority and the rational requirements related to it.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
The question of whether we rely on someone’s beliefs or on their assertions is a complex one that is much discussed. Typically, we have access to testifiers’ beliefs via their utterances. From their more general behavior, we can also obtain clues as to the level of certainty or credence with which they hold those beliefs, even though exact values are, of course, rarely determinable. Moreover, the testifier’s belief does not seem necessary for learning from a speaker. Lackey (2008) demonstrates this with her creationist teacher case. Nevertheless, we will describe cases in terms of reliance on authorities’ beliefs and the credences they give to those beliefs since this is how the debate surrounding epistemic authority proceeds.
Raz restricts this principle to the reasons against the authority’s judgment, so that one can still rationally rely on one’s positive reasons. While this could be reasonable for the case of practical authority, it is implausible in cases of epistemic authority because of the plausibility of the double-counting argument discussed in the next paragraph.
We think of the relationship between the evidence one has about the other person’s epistemic qualities and the belief that she is an authority as that of propositional justification, but we need not commit ourselves to this here and so prefer to use the more neutral description.
Jäger adds that authority status is also time relative. Since this is of limited relevance for our present purposes, we will omit it here for simplicity’s sake.
We will assume in this paper that defeaters must be sufficiently well supported in order to defeat. However, we are not committed to this somewhat controversial (e.g. Bergmann 2006; Alexander 2017) position. Should it turn out that (e.g.) merely believed defeaters are possible, we would be willing to drop the sufficient-reason condition for epistemic authority.
Goldman claims that the truth-values of esoteric propositions are inaccessible to the layperson (Goldman 2001, p. 94). We understand “esoteric” in a weaker sense in that it applies to propositions that laypeople are less reliable in assessing than is the authority, due to their less reliable arsenal of methods. This is meant to leave open the possibility that a layperson can have access to the truth-value of an esoteric proposition. Esotericism is thus a matter of degree.
The general idea of scope-restricted preemption can already be found in Raz (1990, pp. 46–47). However, on Raz’ view the scope-affecting reasons are primarily conventional or institutional facts.
Strictly speaking, it is also not permissible to use E for the assessment of any proposition that is derivable from p. If, e.g., E is preempted from being used as evidence for the assessment of p, then we must not use it as indirect evidence for assessing the proposition p˄(qv¬q), if we inferentially derive this proposition from p.
There are some uses of E where it misleadingly looks as though preemption should hold, but it doesn’t. One example would be a use of evidence E that makes a difference with respect to which attitude to adopt towards p but that does not amount to an assessment of p itself. This may sound strange at first, but cases like this are well known in epistemology: Undercutting defeaters, for example, make a difference regarding what one is to believe, but cannot be used in an assessment of the content of the defeated belief. Another type of apparent counterexample to preemption concerns cases where preemption applies and where one does not use E but rather evidence about psychological states or events that contain or are closely related to E in the assessment of the relevant proposition p. Since evidence about psychological states related to E is not the same as E itself, such cases turn out to be compatible with preemption. We will discuss examples of both kinds of cases in detail later in this article.
Rebutting defeaters, which simply indicate that the defeated belief is false, are supposed to be accommodated by this account because a rebutting defeater is a reason not to hold the defeated belief on any basis at all, including any particular basis (Loughrist 2015, pp. 86–87).
Loughrist discusses such “double-agent” cases, as he calls them, in great detail (Loughrist 2015: 76 ff; 90 ff).
It is important to keep in mind that the defeater that is in play here does not retroactively render reliance on the relevant evidence irrational, which would potentially make it the case that the defeated belief was never rational. Rather, the defeater renders a previously rational belief irrational. An S-defeater, in particular, makes it irrational to rely on the evidence, starting from the point in time when the defeater is obtained. Note that this also affects possible future instances of relying on the relevant evidence: Once the S-defeater is obtained, it is generally irrational to rely on evidence obtained from the source the defeater calls into question.
One might object that the above argument relies on the controversial Principle of Uniqueness (defended by White 2005; attacked by Kelly 2013b). According to Uniqueness, there is exactly one fully rational response to any given body of evidence. One might argue that only under this assumption can we rule out the possibility that a credence that deviates from the authority’s can be rational as well. Without this principle, the authority’s credence and a deviant response that is partly based on the layperson’s assessment may both be fully rational. Uniqueness, however, seems to be highly doubtful as a universal claim (see Kelly 2013b). However, Uniqueness is still true in many cases (Kelly 2013b, pp. 298–299) and the layperson’s credence may lie outside the range of rationally permissible credences in permissive cases. Hence, the layperson must possess evidence for the claim that the case at hand is permissive and some further evidence for the claim that her credence lies in the range of rationally permissible credences in order to acquire a defeater-defeater of the prima facie S-defeater that is constituted by the authority’s credence.
One worry here may be that on our proposal, even conclusive evidence or what the inferior knows can be preempted by authoritative reasons, and you might think that this is a counterintuitive consequence. Notice, however, that preemption is motivated through S-defeat, and S-defeaters can be rationally misleading. On our view, it may be rational to ignore pieces of evidence that are in fact conclusive or known, but this is far from counterintuitive. Consider a case in which the inferior performs some mental calculation in an absolutely impeccable way. She then may confront an epistemic authority about mental math who either is in fact not an expert or is an expert who makes a mistake on this particular occasion. Although the inferior knows her result on the basis of conclusive evidence, she is rationally required to adjust her credence to the in fact inadequate credence of the authority. Admittedly, this is to some extent controversial. Whereas Feldman (2009) and Christensen (2010) claim that higher-order defeaters can even defeat conclusive evidence, Lasonen-Aarnio (2014) disputes this. Williamson (2011) claims that knowledge is indefeasible. We cannot settle this controversy here. We just want to flag that we take higher-order defeat of conclusive evidence and knowledge to be the more plausible option.
The exact way to model cases like this is very complicated because it depends on details in the context, the content of the evidence in play, the strength of the testimonial reason and many other factors. It will thus have to be a subject for future work.
This is, of course, a highly idealized case. If the conflicting authorities assign credences with asymmetric distances from 0.5, a view that favors splitting the difference rather than suspension may seem more appropriate. In this paper, we cannot discuss these more specific questions.
The proponent of TEV might come up with a rival explanation of why it is rational to suspend judgment in the case of conflicting authorities. She might argue that the conflict of authorities indicates that trustworthy judgments about the domain are generally impossible. If even the authorities do not agree, then nobody can have justified beliefs about the domain in question. Thus, one should rationally suspend belief about the target proposition. QED. In response to this line of argument, we must admit that TEV can explain why, in the face of conflicting authorities, suspension may sometimes be rationally required. But this explanation has a certain price. It relies on the assumption that judgments in the domain are generally irrational. In contrast, DP can explain why we must suspend belief when we confront conflicting authorities, even if there is no reason to believe that domain-specific judgments are generally untrustworthy. Suppose, for example, that you observe a specific conflict between authorities and that you know at the same time that there is a broad convergence of authorities on the controversial issue, but you do not know with which side most authorities agree. Intuitively, we must suspend belief even in this case. But in this case, it is hard to see what kind of explanation a defender of TEV could come up with.
This issue is brought up by Jäger (2016, pp. 171, 177, 178), who calls the phenomenon “weak agreement”.
We are grateful to an anonymous referee for articulating this worry in detail.
References
Alexander, D. (2017). Unjustified defeaters. Erkenntnis, 82(4), 891–912.
Bergmann, M. (2006). Justification without awareness: A defense of epistemic externalism. Oxford: Oxford University Press.
Casullo, A. (2003). A priori justification. Oxford: Oxford University Press.
Christensen, D. (2010). Higher-order evidence. Philosophy and Phenomenological Research, 81(1), 185–215.
Dormandy, K. (2018). Epistemic authority: Preemptive reasons or proper basing? Erkenntnis, 83(4), 773–791.
Elga, A. (2007). Reflection and disagreement. Noûs, 41(3), 478–502.
Feldman, R. (2009). Evidentialism, higher-order evidence and disagreement. Episteme, 6(3), 294–312.
Gadamer, H. G. (1975). Truth and method. New York: Seabury Press.
Goldman, A. I. (2001). Experts: Which ones should you trust? Philosophy and Phenomenological Research, 63(1), 85–110.
Grundmann, T. (2011). Defeasibility Theory. In S. Bernecker & D. Pritchard (Eds.), The routledge companion to epistemology (pp. 156–167). New York: Routledge.
Jäger, C. (2016). Epistemic authority. Preemptive Reasons and Understanding. Episteme, 13(2), 167–185.
Kelly, T. (2010). Peer disagreement and higher order evidence. In A. I. Goldman & D. Whitcomb (Eds.), Social epistemology: Essential readings (pp. 183–221). Oxford: Oxford University Press.
Kelly, T. (2013a). Disagreement and the burdens of judgment. In D. Christensen & J. Lackey (Eds.), The epistemology of disagreement: New essays (pp. 31–54). Oxford: Oxford University Press.
Kelly, T. (2013b). Evidence can be permissive. In M. Steup & J. Turri (Eds.), Contemporary debates in epistemology (pp. 298–312). Oxford: Wiley Blackwell.
Keren, A. (2007). Epistemic authority, testimony and the transmission of knowledge. Episteme, 4(3), 368–381.
Keren, A. (2014a). Trust and belief: A preemptive reasons account. Synthese, 191(12), 2593–2615.
Keren, A. (2014b). Zagzebski on authority and preemption in the domain of belief. European Journal for Philosophy of Religion, 6, 61–76.
Lackey, J. (2008). Learning from words: Testimony as a source of knowledge. Oxford: Oxford University Press.
Lackey, J. (2018). Experts and peer disagreement. In M. Benton, J. Hawthorne, & D. Rabinowitz (Eds.), Knowledge, belief, and god: New insights in religious epistemology (pp. 228–245). Oxford: Oxford University Press.
Lasonen-Aarnio, M. (2014). Higher-order evidence and the limits of defeat. Philosophy and Phenomenological Research, 88(2), 314–345.
Loughrist, T. D. (2015). Reasons against belief: A theory of epistemic defeat. Proquest LLC, UMI Dissertation Publishing.
Pollock, J. L. (1986). Contemporary theories of knowledge. Hutchinson: London e.a.
Raz, J. (1988). The morality of freedom. Oxford: Clarendon Press.
Raz, J. (1990). Practical reason and norms (2nd ed.). Oxford: Oxford University Press.
Thune, M. (2010). “Partial Defeaters” and the epistemology of disagreement. Philosophical Quarterly, 60(239), 355–372.
White, R. (2005). Epistemic permissiveness. Philosophical Perspectives, 19(1), 445–459.
Williamson, T. (2011). Improbable knowing. In T. Dougherty (Ed.), Evidentialism and its discontents (pp. 147–164). Oxford: Oxford University Press.
Zagzebski, L. (2012). Epistemic authority: A theory of trust, authority and autonomy in belief. Oxford: Oxford University Press.
Acknowledgements
This paper originates from fully collaborative work of both authors. Early versions were presented at a workshop on social epistemology at the University of Cologne in June 2016, in Gerhard Schurz’ seminar at the University of Duesseldorf in January 2017, at a workshop with Al Casullo at the University of Cologne in May 2017, at a workshop on epistemic authority at the university of Innsbruck in May 2017, at the Bled Epistemology Conference in June 2017 and at a workshop on modern transcendental philosophy at the IUC Dubrovnik in September 2017. Very helpful comments from and extensive discussions with the following colleagues enabled us to work out the final version of this paper: Dominik Balg, Sven Bernecker, Katherine Dormandy, Sandy Goldberg, Frank Hofmann, Joachim Horvath, Christoph Jaeger, Thomas Kelly, Chris Kelp, Arnon Keren, Jens Kipper, Steffen Koch, Jennifer Lackey, Achim Lohmar, Linda Zagzebski, and two anonymous reviewers of Synthese. Finally, we would like to thank the German Research Foundation for supporting our research on the topics of this paper as part of the project Disagreement in Philosophy. Semantic and Epistemic Foundations.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Constantin, J., Grundmann, T. Epistemic authority: preemption through source sensitive defeat. Synthese 197, 4109–4130 (2020). https://doi.org/10.1007/s11229-018-01923-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-018-01923-x