Higher-Order Evidence and Moral Epistemology

Michael Klenk, Michael Klenk

• Publisher: Routledge
• Year: 2020
• ISBN:
  9781000029345
• Edition: 1
• Topic: Philosophy
• Subtopic: Epistemology in Philosophy

This book offers a systematic look at current challenges in moral epistemology through the lens of research on higher-order evidence. Fueled by recent advances in empirical research, higher-order evidence has generated a wealth of insights about the genealogy of moral beliefs. Higher-Order Evidence and Moral Epistemology explores how these insights have an impact on the epistemic status of moral beliefs.

The essays are divided into four thematic sections. Part I addresses higher-order evidence against morality that comes from sources such as disagreement and moral psychology. Part II covers rebuttals of higher-order evidence against morality. The essays in Part III examine the relevance of higher-order evidence for a broader range of phenomena in moral epistemology, for both individuals and groups, including moral testimony and phenomena of practical concern, such as fundamentalist views about moral matters. Finally, Part IV discusses permissible epistemic attitudes regarding a body of moral evidence, including the question of how to determine the permissibility of such attitudes.

This volume is the first to explicitly address the implications of higher-order evidence in moral epistemology. It will be of interest to researchers and advanced graduate students working in epistemology and metaethics.

Part I
Higher-Order Evidence Against Morality

1
Evolutionary Debunking, Self-Defeat and All the Evidence

Silvan Wittwer

1 Introduction¹

Robust moral realism is the metaethical view that there are mind-independent moral facts that are irreducible to natural facts and in principle knowable.² Recently, this view has been criticized heavily on evolutionary grounds. So-called evolutionary debunkers argue that becoming aware of the evolutionary origins of our moral beliefs, realistically construed, undermines their justification.³ While there are various ways of spelling out the underlying epistemological details, an appeal to higher-order evidence of error has recently garnered increased philosophical attention.⁴ More precisely, Tomas Bogardus (2016), Andreas Mogensen (2014, 2017) and – at least on one plausible reconstruction⁵ – Sharon Street (2006) argue that evolutionary considerations provide such evidence, add the view – often called conciliationism⁶ – that higher-order evidence of error defeats justification and thus conclude that evolutionary considerations defeat the justification of our moral beliefs.

In response, moral realists such as Katia Vavova (2014) have objected that this evolutionary debunking argument is self-defeating.⁷ To see how that threat materializes, note that conciliationism is typically motivated by independence, the principle that we should assess higher-order evidence of error with respect to p independently of our original first-order evidence, beliefs or reasoning in support of p.⁸ To respect this principle when assessing evolutionary higher-order evidence of error, the moral realist would need to set aside all of their moral evidence, beliefs and reasoning. But doing so risks setting aside too much to know whether they are genuinely mistaken about morality or not. So the evolutionary debunking argument from earlier seems to defeat itself.

The literature lacks any discussion of whether evolutionary debunkers can handle this self-defeat objection.⁹ My overall aim in this chapter is to argue that they cannot, thus filling that lacuna – and vindicating Vavova’s worry. To achieve my aim, I proceed in two steps. First, I propose a novel, prima facie promising strategy for avoiding self-defeat. Then, I show that evolutionary debunkers face insuperable difficulties trying to successfully implement that strategy. As a result, the evolutionary debunking argument from higher-order evidence of error fails.

What does this strategy (that constitutes the first step of my argument) look like? It consists in realizing that conciliationism isn’t the only prima facie plausible view on the epistemic significance of higher-order evidence of error. In light of that, it would be premature to conclude that the evolutionary debunking arguments from higher-order evidence defeat themselves. Rather, independently compelling alternatives are available – and it remains an issue of active debate which one we should endorse. Further, some of these alternatives appear to have distinctive features that would help avoid self-defeat. In particular, some views (A) reject independence and therefore allow first-order moral evidence into the picture yet (B) still promise to yield the verdict that evolutionary higher-order evidence can defeat the justification of first-order moral beliefs. For example, Thomas Kelly’s (2010) total evidence view fits that bill. According to it, your total evidence determines whether your belief that p is justified or not. The total evidence includes both your first-order evidence (in support of p) and the second-order evidence of peer disagreement (regarding p). Therefore, Kelly’s (2010) view clearly satisfies (A): it rejects independence, affording our first-order moral evidence, beliefs and reasoning a role in determining whether a given belief is justified or not. That amounts to a promising start, giving evolutionary debunkers enough reason to explore whether their argument avoids self-defeat if they accept the total evidence view instead of conciliationism.

However, whether Kelly’s (2010) view satisfies (B) is more complicated, as I show in the second step of my argument. To establish (B), evolutionary debunkers must argue that the total evidence available to the robust moral realist, consisting of their first-order moral evidence and the evolutionary higher-order evidence of error, defeats the justification of their moral beliefs. But, upon reflection, evolutionary debunkers cannot discharge this argumentative burden. The exact reason for their failure depends on the kind of higher-order evidence of error that evolutionary considerations allegedly provide. Debunkers such as Street (2006), for whom evolutionary considerations supply evidence of moral unreliability, struggle with evidential weight. In contrast, debunkers such as Bogardus (2016) and Mogensen (2014, 2017), who construe evolutionary considerations as evidence of moral peer disagreement, are committed to a pair of inconsistent assumptions about evolutionary counterparts. Either way, evolutionary debunkers who appeal to the notion of higher-order evidence of error struggle to implement (B) of my proposed strategy. By implication, their arguments cannot avoid self-defeat.

Here is my plan. Section 2 presents the evolutionary debunking argument from higher-order evidence of error in more detail, while Section 3 turns towards characterizing the self-defeat objection. Then, Section 4 unveils my two-part strategy for dealing with self-defeat. The remainder of the chapter explores whether evolutionary debunkers can make good on that strategy. Section 5 argues that the first part is easy to pull off, but Section 6 raises concerns about the implementation of the second part, having to do with evidential weight and whether evolutionary counterparts qualify as epistemic peers. Section 7 concludes.

2 Evolutionary Debunking and Higher-Order Evidence

Recently, some philosophers have argued that evolutionary theory debunks beliefs in robust mind-independent moral facts. To make their case, some of these evolutionary debunkers rely on the notion of higher-order evidence of error. Despite differences in detail, the arguments put forth by Tomas Bogardus (2016), Andreas Mogensen (2014, 2017) and – at least on Vavova’s (2014) compelling reconstruction – Sharon Street (2006) share the following structure:

• (1) Higher-order evidence of error about your beliefs that p, q, etc. defeats their justification.¹⁰
• (2) Evolutionary considerations provide moral realists with higher-order evidence of error about their moral beliefs.
• (3) Therefore, evolutionary considerations defeat the justification of the moral realists’ moral beliefs.

This argument, if sound, forces moral realists into moral scepticism: even if robust moral facts exist, our beliefs about them aren’t justified. To begin with, the first premise introduces the notion of higher-order evidence of error. Such evidence of error indicates that one suffers from some epistemic malfunction.¹¹ Here are two familiar examples:

Offside Call: In my spare time, I enjoy attending football games with my best friend Julian. We are both equally good at spotting whether a forward is offside or not. Last Sunday, though, we disagreed: while Julian judged that our forward started from an off-side position and the resulting goal was thus irregular, it seemed to me that our forward timed his run well and was onside.¹²

Hypoxia: While climbing the Dufour peak in the Swiss Alps, the weather suddenly turns near the summit. I stop briefly to calculate whether there is enough time to reach the peak and start the climb down before the snow storm hits. After going over my calculations several times, I am rather confident that I should be able to make it. However, I suddenly remember that given the high altitude, I am very likely to suffer from mild hypoxia (or lack of oxygen), which undetectably impairs one’s reasoning, leading to stupid yet fatal mistakes.¹³

In the first case, I receive evidence of peer disagreement: Julian and I are equally good at making offside calls, but disagree about whether our forward was offside or not this time. Since we cannot both be right (but are equally good at making offside calls), one of us must be in error, which might very well be me. In contrast, the second case features evidence of unreliability: hypoxia makes it likely that my reasoning (about time, in this case) is mistaken.¹⁴

The first premise doesn’t just introduce higher-order evidence of error but also articulates a view about its epistemic significance. Conciliationism says that higher-order evidence of error defeats the justification of relevant first-order beliefs.¹⁵ Originally, this view was defended in the context of peer disagreement, recommending that one conciliate (hence its name) upon receiving evidence of peer disagreement, such as in Off-side Call.¹⁶ However, it can easily be generalized, resulting in a view that says that any kind of higher-order evidence of error defeats justification.¹⁷ For instance, it would say that, in Hypoxia, learning of my likely reason-distortion defeats the justification of my belief that I have enough time to reach the Dufour peak and return safely. So conciliationism is a moderate form of scepticism: according to it, higher-order evidence of error defeats justification but only the justification of those first-order beliefs that we have such evidence about.¹⁸

Conciliationism is prima facie attractive.¹⁹ First, it accommodates our intuitions: in both Offside Call and Hypoxia, it seems intuitively appropriate to revise our beliefs in light of the higher-order evidence of error – which conciliationism respects. Second, the view also plausibly explains our intuitions. For instance, given that Julian and I are equally likely to get offside calls right and that there must be a mistake on either his or my part in Offside Call, we both have reason to think that we have made a mistake, which may well be enough to defeat the justification of our relevant beliefs. Similarly, given evidence of my medical condition in Hypoxia, it is significantly more likely that I have made a justification-defeating mistake in my time management. Third, conciliationism follows from a seemingly plausible principle for correctly evaluating evidence. According to independence, we should assess higher-order evidence of error with respect to p independently of our original first-order evidence, beliefs or reasoning in support of p.²⁰ To see why that seems plausible, reconsider Offside Call: once I learn that Julian, an epistemic peer, disagrees with me, it would be intuitively wrong or irrational to dismiss his judgment by depending on my initial perceptual seeming that the forward wasn’t offside. (Similarly, for Hypoxia: sticking with my original reasoning would be epistemically problematic in the face of (significant risk of) an altitude-induced distortion of reasoning.) But once we accept independence, conciliationism straightforwardly follows: if it is rational to bracket one’s first-order evidence and more and thus only the higher-order evidence of error matters, it will defeat the justification of our relevant first-order beliefs.²¹

The second premise states that evolutionary considerations provide moral realists with higher-order evidence of error about their moral beliefs.²² But what kind of evidence is this exactly? There are at least two answers in the literature, both of which work for the argument.

On the one hand, according to Sharon Street (2006), moral realists face a scenario similar to Hypoxia. For her, evolutionary considerations provide moral realists with evidence of moral unreliability.²³ After all, when moral realists reflect on the evolutionary origin of our moral beliefs, they must realize that evolution selects for adaptive, not true, moral beliefs.²⁴ For instance, suppose you believe that you have special moral obligations to your family, based on a corresponding moral intuition. But then you realize that we evolved to survive, not to track mind-independent moral truths, and that it is therefore likely that your belief is false. So moral realists have good reason to think that their moral beliefs have been unreliably formed: they are the upshot of a process that was not designed to get at moral truth. In that way, evolutionary considerations provide moral realists with evidence of moral unreliability.

On the other hand, for Bogardus (2016) and Mogensen (2014, 2017), moral realists find themselves in a situation similar to Offside Call.²⁵ According to them, evolutionary considerations amount to evidence of possible peer disagreement: evidence that moral realists could disagree with their evolutionary counterparts about fundamental moral matters (such as the wrongness of incest or slavery, our obligations towards our children, that ethnicity doesn’t matter to moral standing, etc.). After all, moral realists must realize that, had humans evolved differently, they would hold different moral beliefs now. Suppose moral realists believe that incest is morally wrong, on the basis of a corresponding moral intuition. Their evolutionary counterparts might disagree: since incest did not hamper their reproductive fitness, they don’t believe that it is morally impermissible. Rather, they believe that it is perfectly morally alright, on the basis of their corresponding moral intuition. In that manner, evolutionary considerations amount to evi...

Table of contents

• Cover
• Half Title
• Series Page
• Title
• Copyright
• Contents
• Acknowledgements
• Change in Moral View: Higher-Order Evidence and Moral Epistemology
• PART I Higher-Order Evidence Against Morality
• PART II Rebutting Higher-Order Evidence Against Morality
• PART III Broader Implications of Higher-Order Evidence in Moral Epistemology
• PART IV Permissible Epistemic Attitudes in Response to Higher-Order Evidence in Moral Epistemology
• List of Contributors
• Index