![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
The phenomenon of moral supererogation—action that goes beyond what moral duty requires—is familiar. In this paper, I argue that the concept of supererogation is applicable beyond the moral domain. After an introductory section 1, I outline in section 2 what I take to be the structure of moral supererogation, explaining how it comes to be an authentic normative category. In section 3, I show that there are structurally similar phenomena in other normative domains—those of prudence, etiquette, and the epistemic—and give examples of acts of supererogation of each of these types.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1 An exception is Hedberg [Citation2014], who argues for the existence of epistemic supererogation, as I discuss in section 3.3.
2 Some authors suggest a further necessary condition for supererogation—that the action be praiseworthy. Apparent counterexamples to this include what David Heyd [Citation2015] calls ‘small acts of favour, politeness, consideration and tact’ and, more generally, morally optional behaviour where ‘no particular effort, cost, or risk is involved’. Heyd suggests that there may simply be two conceptual variations of supererogation. See Archer [Citation2016] for argument that supererogatory acts need not be praiseworthy.
3 For characterizations of moral obligation and moral wrongness in terms of the appropriateness of feelings of blame and guilt, see Mill [Citation1861], Gibbard [Citation1990], and Skorupski [Citation2010].
4 For an account of moral supererogation that makes room for such self-regarding duties, see Kawall [Citation2003].
5 An important point is that whether one has responded inadequately to prudential reasons cannot be settled without evaluating the reasons for which one did act. If one acted for sufficiently weighty other-regarding reasons, then one doesn't merit a charge of foolishness, as when one puts one's life at risk in order to save several other people's lives.
6 It is worth noting that Mill himself here avoids the language of obligation, which he reserves for morality, and which he associates with punishment. It is important for Mill that, although being the object of reactions of prudential disapproval is unpleasant, it is nonetheless improper for us to express such disapproval as a way of punishing imprudence [1859: 102]:we may express our distaste, and we may stand aloof from a person as well as from a thing that displeases us; but we shall not therefore feel called on to make his life uncomfortable… He may be to us an object of pity, perhaps of dislike, but not of anger or resentment.
7 Many would judge that ‘mere’ etiquette does not provide reasons for action, or provide reasons for negative responses to falling foul of the deontic standards in question. If falling foul of the code will upset or distress my companions, that will (at least normally) be some reason to comply. But the mere fact that some code of etiquette requires me to act in some way does not obviously provide me with any normative reason to act. Sometimes in fact there are good positive reasons to violate a code of etiquette, perhaps to make vivid that others are taking such norms too seriously, imposing unjustified entry costs to some set of opportunities, such as requiring a dinner suit to attend a student dinner where firms are recruiting.
8 It is for this reason that we might prefer ‘epistemic irrationality’ to ‘epistemic irresponsibility’. ‘Epistemic irrationality’ is not ideal either perhaps, since, as noted below, there are cases of seeming epistemic irrationality where deontic language seems less at home.
9 Note that assertion is just one way of revealing beliefs. Someone who reveals her belief through non-assertoric language or non-verbal behaviour opens herself to the same response.
10 Mere unreliability in avoiding false beliefs may not be sufficient to warrant a charge of epistemic irrationality in our sense, or for us to judge that an epistemic duty has been breached. In a radical sceptical scenario, a brain-in-a-vat may be systematically unreliable in avoiding false beliefs, while adequately responding to evidence, applying correct rules, and so will not merit such charges.
11 It is worth emphasizing that it is our patterns of belief formation, our belief-forming behaviour, the rules or processes that we deploy in forming beliefs, that are the primary focus of such epistemic evaluation. In calling a belief itself irrational, we do so in response to how it was formed, rather than on the basis of its content: Jones may irrationally believe that p, while Smith rationally believes that p.
12 An interesting point of comparison here is with Sinan Dogramaci's ‘epistemic communism’. On his view, too, the function of charges of irrationality is to police belief-forming rules or processes so that we may together reliably acquire true beliefs through testimony. Dogramaci [Citation2012: 522] emphasizes that mere non-normative attributions of unreliability generally won't be enough to motivate those who stray to change their ways. Similarly, I doubt that merely evaluative observations that someone is epistemically suboptimal would have significant motivational impact. Instead, what is required is the special deontic force of an epistemic must that serves to motivate compliance with reliable rules. One must stick to the relevant rules on pain of avoiding the sting of a charge of irrationality. A distinctive aspect of Dogramaci's picture is that such epistemic deontic language is used primarily to secure coordination in belief-forming rules or processes. We call someone ‘irrational’ not just when they adopt unreliable rules, but, more generally, when they adopt idiosyncratic rules. When someone forms beliefs on the basis of reliable rules that are idiosyncratic (perhaps because they are counter-intuitive), we cannot safely trust her testimony since we need to check for ourselves that the rules are reliable. By contrast, when someone deploys belief-forming rules that we ourselves would have used, we can treat her as an epistemic surrogate—she forms the same beliefs that we would have formed on the basis of the same evidence. On Dogramaci's view, then, our epistemic deontic language is used to police those in our epistemic community not just into reliable rules, but, more specifically, into a common set of reliable rules. See Dogramaci [Citation2012, Citation2015].
13 Compare Dogramaci [Citation2012: 782].
14 Hedberg [2014: 3632] questions whether there is any duty to strive to increase our stock of true beliefs.
15 There may be some sense in which we exclude someone who has these other epistemic vices. We may leave him out of our discussions, because he ‘brings nothing to the table’. But the exclusion underlying epistemic duty is more akin to ignoring someone's assertions.
16 Epistemic over-caution is rare, compared to epistemic rashness. Epistemic over-caution should also be distinguished from a closely related practical fault. We may issue a practical criticism of over-caution towards those who refuse to act in the face of uncertainty, who misweigh the possibility of future benefits of learning more against the costs of present inaction. In the case of global warming, for instance, it is foolhardy to insist on greater evidence before deciding to act. But this practical vice of over-caution is distinct from the epistemic vice of over-caution (over-reluctance to assent to a proposition in the face of available evidence).
17 For our purposes, we need not endorse some particular conception of epistemic goodness. (More than one such conception may be useful, in any case). So long as some such conception is in good order, then, there will be a phenomenon of epistemically supererogatory behaviour that fits our model: behaviour that is epistemically better than the minimum that is required for meeting one's epistemic duties, where such duties are characterised by reference to warranted exclusion from the epistemic community.
18 Epistemically supererogatory beliefs, then, are those formed on the basis of such epistemically supererogatory behaviour—e.g. those beliefs acquired by doing epistemically optional reading of encyclopaedias.
19 This paper was presented at the Universities of Basel, Reading, and York. I am very grateful to the audiences for very useful discussion. For extremely helpful feedback on earlier drafts of the paper, I would like to thank Sebastian Becker, Iason Gabriel, Felix Pinkert, Enzo Rossi, Richard Rowland, Joe Slater, Justin Snedegar, Jesse Tomalty, Jonathan Way, Daniel Whiting, and referees and editors for this journal.