Moral Deference, Moral Assertion, and Pragmatics

Abstract

In this paper, I offer a novel defense of moderate pessimism about moral deference, i.e., the view that we have pro tanto reason to avoid moral deference. I argue that moral deference fails to give us the epistemic credentials to satisfy plausible norms of moral assertion. I then argue that moral assertions made solely on the basis of deferential moral beliefs violate a plausible epistemic and moral norm against withholding information that one knows, has evidence, or ought to believe will importantly affect another person’s deliberation. Finally, I argue that not only does moral deference fail to put the audience in a good epistemic position it also puts the audience in a bad epistemic and moral position. First, there is a tight connection between outright believing something and being disposed to assert it and so deferential moral beliefs often motivate people to assert something that they don’t have the epistemic credentials to properly assertion. Second, there will often be moral reasons to make assertions—even based on deferential moral beliefs. These assertions, while all-things-considered permissible, will be epistemic impermissible and involve violating a moral norm.

1 Introduction

Many philosophers think that there is something amiss with moral deference, i.e., forming a moral belief solely on the basis of another’s say-so. However, there is much disagreement over whether there is anything actually (epistemically or morally) bad about it. The two most popular views are moderate pessimism and moderate optimism. According to moderate pessimists, we have pro tanto reason (of some kind) to not engage in moral deference but it is often outweighed by other pro tanto reasons.¹ On the other hand, moderate optimists deny that we have pro tanto reason to not engage in moral deference but nonetheless think that there is something sub-optimal about it.² In what follows, I refer to these positions simply as “pessimism” and “optimism.”

Despite their disagreement, most pessimists and optimists endorse some version of the following:

Incompatibility: Moral deference (as a belief-formation method) is incompatible with acquiring some target epistemic state (e.g., moral understanding) and therefore fails to put one is a good position for certain moral achievements (e.g., acting with moral worth).

The problem with the existing defenses of Incompatibility is the supposed incompatibility is just that moral deference does not transmit or directly result in the relevant target epistemic states (e.g., moral understanding). However, there is no sense in which using moral deference prevents or inhibits the gaining of the target epistemic states at some later time.³ In fact, believing that p (regardless of the belief-formation method) makes it rational to try to seek out the target epistemic states, e.g., understanding why p.

Because the incompatibility that most authors argue for is so weak, it is implausible that we have pro tanto reason to not use moral deference. At best, the fact that moral deference doesn’t yield understanding (or other target epistemic states) only warrants the claim that it is sub-optimal, i.e., it only supports optimism. Thus, if pessimism is to remain a reasonable alternative to optimism, pessimists will have to argue for a stronger incompatibility between moral deference and some target epistemic state. Call this the Incompatibility Challenge.

In this paper, I offer a novel defense of moderate pessimism that can answer the Incompatibility Challenge. In particular, I argue there is something both epistemically and morally pro tanto bad about engaging in moral deference. First, I argue that moral deference doesn’t yield the epistemic credentials needed to satisfy plausible accounts of the norm of moral assertion. This means that one cannot epistemically permissibly assert a moral proposition solely on the basis of a deferential belief. Second, I argue that when one makes a moral assertion, one entitles one’s audience to believe falsehoods about the asserter’s epistemic credentials concerning the content of her assertion. Third, and relatedly, I argue that the asserter knows, has good evidence, or ought to believe that if her audience knew that she lacked these epistemic credentials, they would not defer to her. Finally, I argue that there is a tight connection between outright believing that p and being disposed to assert that p. Thus, when one forms a deferential moral belief that p, one enters oneself into a tension between one’s dispositions (to assert that p) and what is epistemically and morally permissible (i.e., to not assert that p).

The plan is as follows. In §2, I detail the kind of deference and assertion with which I’m concerned. In §3, I discuss a few plausible accounts of the norm of moral assertion and argue that one cannot satisfy these norms by asserting solely on the basis of a deferential moral belief. In §4, I argue that deferential moral assertion violates an additional epistemic/moral norm. In §5, I detail the link between belief and assertion and argue that it creates a tension in a person when they have a deferential moral belief. In §6, I consider an important objection to account of what is bad with moral deference. In §7, I briefly conclude.

2 Moral Deference and Pure Moral Content

2.1 Moral Deference⁴

I’m only concerned with the following kind of moral deference. First, the deference is purely moral, i.e., the deferrer only defers about thin moral content. For example: right/wrong, permissible/impermissible, good/bad, obligatory/optional, etc. Second, the deferrer forms a full moral belief completely on the basis of the moral expert’s (or her epistemic peers’) testimony. By this I mean: (a) the deferrer forms an outright or full belief on the basis of moral deference and (b) that outright belief is only/completely based on that another’s say-so. Thus, the kind of deference I’m interested in is pure, full, and complete.

The reason for focusing on this narrow set of propositions is to eliminate possible confounding factors. First, the testifier is restricted to an epistemic superior of the deferrer because otherwise the oddness of deference could be explained by the fact that the deferrer is deferring to an epistemic peer or inferior.⁵ Second, the deferrer’s belief is solely based on the testifier’s say-so to ensure that our assessment of her belief is not based on any other grounds of her belief. Third, the deference only concerns purely moral propositions because impure moral propositions mention the grounds of the moral property, e.g., “David’s action was wrong because it constituted failing to maximize happiness.” The problem with providing this additional content is that the deferrer might use this content as another basis of her belief and thus not completely base her belief on the testifier’s word. Or, she might infer from (1) the claim that the action constituted failing to maximize happiness and (2) her background knowledge that failing to maximize happiness makes actions wrong to (3) David’s action is wrong. In either case, her belief is not completely based on the testifier’s say-so. Fourth, the testifier’s belief concerns thin moral content because thick moral content entails descriptive content that the deferrer could use to infer the moral status of the acts, events, persons, etc. under consideration. Finally, the deference concerns outright belief as opposed to credences because the former settle a person’s view and make further inquiry unjustified in many cases.⁶

2.2 Pure Moral Assertion

I will primarily be concerned with flat-out, pure, moral assertions. These assertions are purely moral because all they do is predicate a thin moral property of some action, event, state of affairs, mental state, or person. These assertions are flat-out in that they are unhedged. When one hedges one’s assertion that p one weakens one’s commitment to p. Exactly what this commitment consists in need not concern us here, but options include whether one is committed to defending p or whether one is appropriately liable to blame if not-p. One can hedge an assertion by mentioning that one has a weak attitude toward the proposition asserted (e.g., “I believe Chetan moved to Chicago” or “Chetan moved to Chicago, I think.”) One can also hedge an assertion by mentioning that one has weaker epistemic credentials concerning p than are required for appropriately flat-out asserting (e.g., “From what I hear, Chetan moved to Chicago.”).⁷

3 Accounts of the Norm of Moral Assertion

3.1 The Norm of Assertion

What kind of norm is the norm of assertion? As I will understand it, it is an epistemic requirement that an asserter must meet in order for her assertion to be epistemically proper or permissible. It is certainly true that assertion, as a kind of act, is governed by many kinds of norms, e.g., moral, rational, social, and so on. However, assertions are also common sources of beliefs for others and so they can have epistemically-relevant properties, e.g., reliability and truthfulness.

There are a variety of differences amongst proposed norms of assertion. First, the epistemic norms that I will mention are meant to be necessary to assertion. In other words, assertions are always governed by this norm. Second, the norms I will mention are defeasible, i.e., they can be overridden by other norms that govern assertion—including other epistemic norms. That being said, if the norm of assertion is overridden by a non-epistemic norm and the asserter doesn’t satisfy the norm of assertion, her assertion is epistemically impermissible, even if it is morally or all-things-considered permissible.⁸

3.2 The Norm of Assertion Vs the Norm of Moral Assertion

I follow Simion (2018) in thinking that the norm of moral assertion is stronger than the norm of garden-variety assertion. There are three reasons to think this. First, there is a tight connection between one’s moral beliefs and various of one’s non-cognitive states (e.g., motivations, desires, emotions, etc.). This means that influencing one’s moral beliefs means having a larger impact on them than influencing one of their many non-moral beliefs. Second, there is the practical/moral import that moral assertions have. When someone trusts my moral assertion it will likely play a role in how they behave and whether they perform wrong, bad, or blameworthy actions.⁹ This is often not the case in non-moral assertion.¹⁰ There is a moral risk in trusting moral assertions. Finally, some think that the goal of moral inquiry is moral understanding. If the purpose of assertion is to help audiences reach the goal of the inquiry they are engaged, then it looks like moral asserters are on the hook for helping their audiences gain moral understanding. However, you might think that in order to help an audience gain moral understanding, one must not only know or justifiably believe that p, one must also understand why p or be able to provide an explanation of why p. Thus, it strikes me as plausible that an asserter must be better epistemically placed to properly assert moral propositions than they must be to assert garden-variety ones.¹¹

3.3 The Norm of Moral Assertion

In order to defend some version of moderate pessimism about moral deference, I need not assume any particular account of the norm of moral assertion. Rather, I just need one which cannot be met by a single instance of moral deference. Thus, I can rely on a norm of moral assertion that has at least one of the following conditions: (1) The asserter must understand why p, e.g., the asserter must have some grasp or appreciation of some of the morally relevant considerations are that bear on p and one must also grasp or appreciate how they bear on p, (2) the asserter must always immediately accompany his moral assertion with an explanation of why p, (3) the asserter must be capable of providing an explanation of why p, (4) the asserter must have used her own evaluative capacities to form her judgment that p, or (5) the asserter knows that p, but moral knowledge requires some degree of moral understanding or a grasp of the relationship between morally relevant considerations for p and p itself.¹²

While I prefer a norm that requires something like (1) and (3), others, such as Simion (2018) prefer something like (2). Simion (2018) defends the following norm of moral assertion:

The Explanation Proffering Norm of Moral Assertion (EPNMA): At a context C, one’s moral assertion that p is epistemically permissible only if (1) one knows that p and (2) one’s assertion is accompanied by a C-appropriate explanation why p.¹³

She includes a contextual component because what counts as an explanation or understanding might be context-sensitive. Moreover, certain contextual matters will determine how elaborate or detailed the explanation needs to be in order to produce understanding in one’s audience, e.g., the audience’s previous knowledge and intelligence, the general assumptions about morality at the context, the difficulty of the topic, how controversial the topic is, how much morally hinges on the answer, the urgency of the situation, and so on.

I prefer the following norm of moral assertion:

The Understanding Norm of Moral Assertion (UNMA): One’s moral assertion that p is epistemically permissible only if (1) one meets the norm of garden-variety assertion concerning p on the basis of understanding why p and (2) one understand why p enough that one can offer an explanation of why p that a morally mature agent can grasp.

I’m remaining neutral on what the norm of garden-variety assertions is.¹⁴ As I see it, the norm of moral assertion is just whatever the norm of garden-variety assertion is plus conditions unique to moral assertion.¹⁵ Both of these norms are well-motivated by the intuitive goal of moral inquiry: moral understanding.¹⁶ Notice that both involve moral explanation and not just mere assertion plus some epistemic state (e.g., knowledge or understanding). This is because understanding does not seem to be the kind of epistemic state that can be transmitted or generated by assertion.

I think UNMA is more plausible than EPNMA because it avoids important problems that the latter faces. First, EPNMA seems to entail that making a flat-out moral assertion is never epistemically permissible. Rather, one must always follow a moral assertion with a context-appropriate explanation of why the moral assertion is true. However, there are clear cases in which unaccompanied moral assertions are intuitively epistemically appropriate. For example:

Antebellum Slavery: Michael is the son of a slave owner in the antebellum south of the United States. He has lived on the same estate as slaves his entire life. Now 21, Michael’s long-lived doubts about the morality of slavery have come to a head and he now knows that slavery is wrong and he also grasps why. During an argument with father, he says, “How could you be so blind? Slavery is clearly wrong!”

Michael’s moral assertion that, “Slavery is clearly wrong!” strikes me as epistemically permissible.

Second, EPNMA requires that the asserter know a considerable amount about her audience because she needs to be able to provide a context-sensitive explanation. For example, in order to make an epistemically proper moral assertion to H, a speaker, S, must know H’s previous moral knowledge, her intelligence, and perhaps what morally hinges on the truth or falsity of the assertion. This severely restricts to whom one can make moral assertions. Moreover, this makes the norm of moral assertion disanalogous to the norm of garden variety assertion. Notice that almost none of the accounts of the norm of garden-variety assertion requires that the speaker know anything about the speaker.¹⁷ UNMA has none of these problems and thus is preferable to EPNMA.

3.4 The Norm of Moral Assertion and Moral Deference

The problem with moral deference is that it is epistemically impermissible to make moral assertions on the basis of deferential moral beliefs. This is because deferential moral beliefs are not sufficient for acquiring understanding of why some moral claim is true or for being able to provide an explanation (context-appropriate or otherwise) of why the content of one’s assertion is true.¹⁸ Coming to have moral understanding will most often require having some contact or access to the morally relevant properties of an action, event, state of affairs, etc. This contact can be through imagination (e.g., thought experiments), perception (e.g., witnessing the event in question), or reasoning (e.g., analogizing one action to another). And moral deference does not provide such contact or access. Thus, moral deference fails to put a person in a position to (epistemically) permissibly assert the contents of their deferential moral beliefs.

4 Deferential Moral Assertions and False Implicature

Asserting on the basis of moral deference not only violates plausible accounts of the norm of moral assertion, but it often also violates other epistemic and moral norms. One such norm is the following:

Deal-breaker: S has pro tanto reason to not assert that p to H if the following conditions hold:

• 1. a. S knows (has good evidence or ought to believe) that S’s asserting that p will entitle H to believe that S or S’s assertion has feature f, where S knows that neither S nor S’s assertion has f; and

• 2. b. S knows (or has good reason to believe or ought to believe) that if H knew that S or S’s assertion lacked f, then H would not either partly or completely defer to S about p, i.e., H would not base H’s belief that p either partly or solely on the basis of S’s assertion that p.

The basic idea behind this norm is that one shouldn’t assert that p to someone that one knows (or has good evidence or ought to believe) will not trust that assertion if they knew one’s epistemic credential concerning p. Importantly, there are two parts to Deal-Breaker and I’m claiming that one has a pro tanto reason to avoid making assertions only if one’s assertion satisfies both conditions.

The first condition concerns misleading implicatures. Much of what we say entitles our interlocutors to draw certain conclusions, e.g., that we believe what we’re saying or that we think what we’re saying is relevant to the current conversation. Sometimes, however, what we entitle others to believe will be false. While it’s preferable to avoid entitling others to form false beliefs, oftentimes these false beliefs will be practically insufficient. In particular, the interlocutor will never have to act on this false belief or, if she did, it would have no effect on what she might do. For example, imagine that Simon is responsible for setting up tables and chairs for a conference reception. He wants to know if the tables have been delivered to the reception room. He asks Carol if the tables are in the reception room and she answers yes. Given that, in the vast majority of cases, when someone asserts that something is in another room they know this by seeing it themselves, Carol’s assertion entitles Simon to believe that Carol saw for herself that the tables were in the other room. However, let us grant, that Carol didn’t see the tables in the other room, but rather learned this via Nancy’s testimony. Thus, Carol’s assertion satisfies the first condition.¹⁹

Does satisfying the first condition suffice for giving one a pro tanto reason to not assert? I think the answer is that sometimes it is sufficient for generating a pro tanto reason to not assert. Entitling people to believe falsehoods is bad, from an epistemic point of view. However, most of the time the beliefs that the assertion will entitle audiences to form will not be epistemically important to the audience. That is, it is not the kind of belief that the audience will use as a basis for other beliefs or a reason to drop conflicting beliefs and so on. Thus, while it might be false, it will not often “infect” other of one’s beliefs. So, one has only a weak epistemic reason to not make assertions with misleading implicatures.

However, I think that Deal-Breaker, read as a moral principle, more plausibly grounds a pro tanto reason to not make certain assertions. If one’s assertion satisfies the first condition, then one is either intentionally, knowingly, or negligently deceiving another person. When one violates Deal-Breaker, one (a) intentionally presents another person with misleading evidence, or (b) presents another person with evidence that one foresees will mislead them, and (c) presents another person with evidence that one should foresee will mislead them. But it is pro tanto wrong to intentionally, knowingly, or negligently mislead another person. So, violating Deal-Breaker is pro tanto wrong. However, there will be plenty of times when this wrong is defeated. This is because all communication comes with the risk of misleading someone and people freely and knowingly engage in communicative practices. Likewise, most of the time, the proposition that the audience is misled about is not practically important, i.e., it will not affect the audience’s actions nor will they care about having a false belief about the matter.

For this reason, I’m not concerned with claims that only satisfy the first condition of Deal-Breaker, but only with those that satisfy both conditions. The second condition concerns not disclosing information that one knows (or has good evidence or ought to believe) will change how the audience responds to one’s assertion. In particular, one knows (or has good evidence or ought to believe) that the audience will either not believe the content of one’s assertion or will only treat one’s assertion as a partial basis of belief. Satisfying this condition, I think is sufficient for generating a pro tanto reason to not assert (without disclosure). This is because in withholding information that one knows (or has good evidence or ought to believe) will importantly affect how the audience doxastically responds to one’s assertion constitutes culpably exerting an inappropriate amount of control over the audience’s deliberation. Here the asserter is intentionally (or recklessly or negligently) controlling the audience’s decision about what to believe. Thus, it is an undercutting of a person’s epistemic agency.

In not disclosing the relevant information, the asserter is, at best, showing a lack of respect or regard for the audience’s epistemic standards or values. At worst, the asserter is engaging in a kind of epistemic paternalism in which she thinks her epistemic standards are superior to those of her audience. In both cases, the asserter knows that the audience would treat the undisclosed information as a deal-breaker for belief, but is nonetheless acting as if this doesn’t provide her with a reason to disclose the information.

But this is not the full extent of wrong that the asserter commits. Recall that the assertion entitles the audience to believe the claim that, were she to know it is false, would prevent her from either fully to partly trusting the speaker’s assertion. So not only is the asserter responsible for putting the false belief in the audience’s head, she’s also responsible for then using that false belief to get the audience to believe something the asserter knows she likely wouldn’t otherwise believe.

One might think object to Deal-Breaker by noting that there are cases in which an assertion violates Deal-Breaker but where making that assertion doesn’t seem pro tanto wrong.²⁰ For example, consider the following case from Lackey (2008):

Creationist Teacher: Stella is a devoutly Christian fourth grade teacher, and her religious beliefs are grounded in a deep faith that she has had since she was a very young child. Part of this faith includes a belief in the truth of creationism and, accordingly, a belief in the falsity of evolutionary theory. Despite this, Stella fully recognizes that there is an overwhelming amount of scientific evidence against both of these beliefs. Indeed, she readily admits that she is not basing her own commitment to creationism on evidence at all but, rather, on the personal faith that she has in an all-powerful Creator. Because of this, Stella does not think that religion is something that she should impose on those around her, and this is especially true with respect to her fourth-grade students. Instead, she regards her duty as a teacher to include presenting material that is best supported by the available evidence, which clearly includes the truth of evolutionary theory. As a result, while presenting her biology lesson today, Stella asserts to her students, “Modern-day Homo sapiens evolved from Homo erectus,” though she herself neither believes nor knows this proposition to be true.²¹

One might object that Stella’s assertion violates Deal-Breaker but is not pro tanto wrong. In other words, the claim is that: (1) Stella knows (has good evidence or ought to believe) that her utterance, “Modern-day Homo sapiens evolved from Homo erectus” entitles her students to believe that she knows this claim when she knows she doesn’t know it and (2) Stella knows (has good evidence or ought to believe) that if her students knew that she didn’t know that “Modern-day Homo sapiens evolved from Homo erectus,” then they would not either partly or solely base their belief in the above proposition on Stella’s assertion that p.

I think the best response to this objection draws on Milić (2017). According to Milić, when Stella utters, “Modern-day Homo sapiens evolved from Homo erectus,” she is either uttering the hedged claim, “According to the best available evidence, modern-day Homo sapiens evolved from Homo erectus” or she is performing a speech act similar to, but distinct from, assertion.²² Thus, I don’t think this objection works. First, I think that what Stella knows (has good evidence of or ought to believe) is that she entitles the students to believe that she knows that “According to the best available evidence, modern-day Homo sapiens evolved from Homo erectus”) and not that she knows that, “Modern-day Homo sapiens evolved from Homo erectus.” Second, I don’t think Stella knows (or has good evidence) that if she students knew that she didn’t know that, “Modern-day Homo sapiens evolved from Homo erectus,” they would not base their belief in this claim either partly or fully. This is especially true if they know that she knows that the best available evidence supports this claim.

Finally, one might worry that Deal-Breaker overgeneralizes in a different way by allowing that misleading an audience about any feature can matter to whether a person can permissibly make an assertion.²³ For example, imagine the following case:

London Calling: Amia knows that her assertion, “Andrew Wiles solved Fermat’s Last Theorem,” entitles Tim to believe that she is from London (because of her accent) and she knows that she’s not from London. Imagine further that she knows that were Tim to know that she is not from London he wouldn’t believe her (because he only trusts Londoners).²⁴

One might think that it is intuitive that it would not be wrong for Amia to make the above assertion. The idea behind this objection might be that it is bad to allow features that are epistemically-irrelevant to the truth or belief-worthiness of the assertion matter to whether a person can permissibly make an assertion.

I still have the intuition that Amia has a pro tanto reason to not make the above assertion to Tim. After all, what justifying reason could she have to make such an assertion to him? Perhaps Amia should assert all-things-considered, because she really values telling others facts about mathematics. But even this is odd. It looks like Amia dismissing Tim’s epistemic preference for Londoners just so that she can act in accordance with a quirky value that she has. Of course, Amia might have a very good reason to make the above assertion such that it outweighs her reason to not deceive Tim, e.g., making the above assertion to Tim is the only to prevent Tim from hitting someone. But this is consistent with Deal-Breaker, because, in that case, the pro tanto reason against asserting is defeated.

If one is unconvinced that Amia has a pro tanto reason to not make the above assertion, then one can read Deal-Breaker as only about misleading people about features that are actually or reasonably believed to be relevant to the truth or belief-worthiness of the assertion. Call this principle Deal-Breaker*. Amia’s doesn’t violate Deal-Breaker* by making the assertion above. This is because whether or not she is from London is irrelevant to the truth or belief-worthiness of the assertion, “Andrew Wiles solved Fermat’s Last Theorem.” However, had the content of her assertion been directions to Piccadilly Circus, she would have violated Deal-Breaker*, because her being from London is relevant to the truth or belief-worthiness of the assertion. However, I take it that in this latter case, she intuitively does have a pro tanto reason to not violate Deal-Breaker*.

4.1 Deal-Breaker and Moral Assertion

How does making a moral assertion solely on the basis of moral deference violate Deal-Breaker? First, when I make a pure moral assertion, it entitles my audience to believe that I have judged this proposition to be true myself (e.g., using my own moral knowledge, moral understanding, or sensibilities (e.g., intuitions, gut feelings, etc.)), and/or that I likely know some of the consideration in favor of it, and/or that I can explain why it’s true. But, all of this is false in the case of deferential moral beliefs. Second, asserters know, have evidence, or ought to believe that were their audiences to know that they lack the above epistemic credentials, they either wouldn’t fully or wouldn’t partly defer to the asserter.

Why should we believe that pure moral assertions epistemically entitle an audience to believe the aforementioned proposition? One reason is that one of the accounts of the norm of moral assertion mentioned above (or one close it) is true and implicitly recognized by people. Another reason, which is compatible with the second, is that the following conjunctions are true:

• (1) When people assert that p they represent themselves as believing that p or people’s assertions most often accurately report or express their beliefs; and

• (2) People most often form their moral beliefs on the basis of making their own judgments, i.e., by using their own evaluative capacities; and

• (3) (1) and (2) are common knowledge.²⁵

• (4) When people assert that p without qualification, they represent themselves as being in an epistemically typical position concerning p (i.e., they represent themselves as being able to provide non-testimonial reasons in support of p); and

• (5) Most of the time, when a person makes a flat-out moral assertion that p, that person is capable of providing non-testimonial reasons in support of p other than the testimony of a third party; and

• (6) (4) and (5) are common knowledge.

• (7) When people assert that p they most often accurately represent themselves as believing that p.

• (8) Most of the time, when a person has a moral belief that p, that person knows what some of the morally relevant considerations that bear on p are and grasps (to some degree) how those consideration bear on p; and

• (9) (7) and (8) are common knowledge.

From (1)–(2) it follows that it is likely that when S asserts that p, where p is a moral assertion, S believes that p and formed this belief on the basis of S's own judgment. From (3), it follows that the audience is likely to know (1)–(2) and thus is justified in inferring that the speaker formed her moral belief on the basis of exercising her own moral judgment. Moreover, if the audience knows that (3), it follows that the audience can justifiably infer that the speaker likely knows (1)–(2) and if (1)–(2) weren’t true of his case, he would have said so. The same story can be told about (4)–(6) and (7)–(9).

The false implicatures involved here are conversational implicatures. According to Grice (1989), conversations have the purpose of exchanging information in a maximally effective way. If this is the purpose of conversations, then you should “make your conversational contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged.”²⁶ Grice (1989) calls this the Cooperative Principle. Following this principle involves following four maxims. For our purposes, the most important maxim is the maxim of quantity: Make your [conversational] contribution as informative as is required (for the current purposes of the exchange).²⁷ When one asserts during a conversation, one normally represents oneself as being cooperative and thus as following all of Grice’s maxims.

In the case of making moral assertions, one represents oneself as following the maxim of quantity and thereby implicates that one’s assertion is as informative as is required for the current purposes of the exchange. In the cases of implicature I have in mind, the purpose of the conversational exchange is to figure out whether the audience should believe the content of one’s moral assertion. When one makes a moral assertion, but doesn’t mention that one doesn’t understand why the proposition is true, or that one didn’t form the belief based on one’s own evaluative capacities, or that one can’t offer an explanation of why the proposition is true, one implicates that one does have these epistemic credentials concerning the moral proposition in question. Why? Because otherwise one’s moral assertion wouldn’t be as is informative as required for the purpose of figuring out if the audience should believe the moral claim the asserter utters. After all, what epistemic credentials the asserter has concerning the moral proposition that is asserted are relevant to whether the audience should believe the content of the moral assertion. By representing one’s assertion as being as informative as is required for the purposes of the conversation, one implicates that there is nothing out of the ordinary about what epistemic credentials one has concerning the moral proposition asserted. It is this implication that entitles the audience to form certain false beliefs about the asserter’s epistemic credentials.

Why should we believe that most audiences either wouldn’t fully or wouldn’t partly defer to a moral asserter if they knew that asserter lacked the above-mentioned epistemic credentials? First, full moral deference is rare in real life and so regardless of whether the asserter has the proper epistemic credentials, it’s unlikely that their audiences would fully defer anyway. Second, in the case of partial deference, I think many audiences wouldn’t even partly defer if they found out that the asserter lacked the aforementioned epistemic credentials.

First, it seems like audiences want to know why some moral claim is true before they believe it. As Coady (1992) notes, “When someone tells me that what I am proposing to do is immoral, I do not react by asking for his credentials but for his reasons.”²⁸ If one cannot offer an explanation of why the moral claim is true, and one’s audience knows this, then they are unlikely to even partly defer to one. Second, following one’s moral assertion with an admission that one lacks any of the above epistemic credentials sounds like one is undercutting one’s epistemic authority to assert that claim. For example, consider the following conjunctions:

• 1. “You’re morally required to donate to charity, but I don’t understand why you’re morally required to donate to charity.”

• 2. “It’s morally wrong to donate money to hate groups, but I don’t know what any of the moral considerations are for thinking that.”

• 3. “It’s morally wrong to deadname people, but I can’t explain why it’s true.”

Thus, in clarifying their epistemic position, deferential asserters intuitively undercut their epistemic authority to be making the particular moral assertions they make. This, I think, gives us good reason to think that audiences would not even partly defer to deferential asserters if they knew that the asserter lacked the above epistemic credentials. Moreover, just ask yourself, “Would I even partly defer to someone about a moral claim when that person doesn’t understand why it’s true and can’t tell me any of the moral considerations that bear on its truth?” I think the answer for most of us would be, “No.”

Thus, I think we have good reason to think that making moral assertions solely on the basis of deferential moral beliefs violates Deal-Breaker and thus we have pro tanto reason to not assert moral claim solely on the basis of deferential moral beliefs.

4.2 Objections

A natural objection here is that the deferential asserter could also defer about why the moral proposition is true and thereby be able to give her advisees non-testimonial evidence. The thought here is a good one. This way, the deferential asserter would not mislead their audience about being able to provide an explanation of why p.

First, even in this case, the asserter would be misleading that audience that she formed her moral belief on the basis of her own judgment and that she grasps (to some degree) how some of the moral considerations that bear on p actually bear on p. Second, if she were to assert “p because q,” she would be misleading her audience about something else, i.e., that she bases her belief that p on q. But this is false. She deferred about this explanation. Moreover, her assertion that “p because q” will be insincere because she doesn’t believe p on the basis of q, she just believes the proposition “p because q.”²⁹

Finally, it will not do for the deferential advisor to base her belief that p on q, because then she will no longer be asserting on the basis of a deferential belief that p and I’m only arguing that there is something epistemically and morally improper about asserting a pure moral proposition on the basis of a deferential moral belief.

A second objection goes as follows: The asserter can simply cancel any misleading communication, e.g.:

“It’s wrong to spend thousands of dollars on luxury clothing instead of sending that money to charities, but I have no understanding of why this is true.”

First, this might cancel the misleading communication, but that does not change the fact that one could not properly make flat-out pure moral assertions.³⁰ This assertion is tantamount to a hedged assertion. Second, this assertion strikes me as infelicitous. Unlike with a conversational implicature, it doesn’t look like I can easily cancel the non-assertoric content concerning my understanding of why my assertion is true.

It also seems plausible that once the advisee finds out that the asserter did not use her own moral understanding, knowledge, or sensibilities to form this judgment, they are much less likely to take treat the utterance as an assertion, i.e., as licensing them to stop inquiring into whether p. The audience might increase their credence in the proposition, but it strikes me as reasonable that they should not form an outright belief on the basis of this person’s assertion.

The problem with flat-out asserting a moral proposition solely on the basis of deference and then canceling the misleading evidence is that there is a far less misleading way of communicating the same information that should be obvious to the asserter. Instead of asserting, “It is wrong to commit a microaggression. But I didn’t judge that for myself nor do I know why it’s true. Rather, a reliable advisor told me,” one could simply assert, “A reliable advisor told me that it is wrong to commit a microaggression and I believe them.” For example, the first wording seems to violate the Gricean maxim of manner according to which one ought to speak in an orderly and non-ambiguous manner. In uttering this in a conversation, the audience might be entitled to believe that the speaker is trying to implicate something by violating the maxim of manner. For example, the audience would be rational in asking themselves, “why didn’t the speaker just tell me that he heard someone else say that microaggressions are wrong?” and then try to infer some conversational implicature from the speaker’s choice of wording.

5 The Incompatibility Challenge

5.1 Restating the Challenge

Recall that the Incompatibility Challenge for pessimists is to provide an account of how moral deference and some epistemic state or moral achievement are incompatible such that one has a pro tanto reason to avoid moral deference. Importantly, the incompatibility can’t just be that moral deference doesn’t transmit or directly result in some target epistemic state. After all, this is compatible with a deferrer immediately reaching the target epistemic state right after deferring. Thus far, I’ve just argued that moral deference fails to put one in a position to epistemically permissibly assert the content of one’s deferential belief. Thus, I must say more if I’m to avoid falling prey to the Incompatibility Challenge.

5.2 Meeting the Challenge

What makes deferential moral beliefs so bad is not that they fail to put someone in some good epistemic or moral position, it’s that they do put someone in a bad epistemic or moral position. To understand why, we need to consider a particular aspect of outright belief. What I have in mind is that many think that there is a tight connection between outright believing a proposition and being disposed to assert that proposition in some range of circumstances.

Many will find the following link plausible:

Belief-Assertion Link: If you (outright) believe that p, then you are disposed (to some degree) to assert that p in some range of circumstances—all else being equal.

There are at least two ways of explaining the link between outright believing that p and being disposed to assert that p. On a weaker reading of Belief-Assertion Link, the connection between the two is merely contingent, though nonetheless strong. For example, it may just be a fact about our psychology that we are disposed to assert whatever we outright believe when it is conversationally relevant to do so. On a stronger reading, there is a conceptual link between the two. In other words, part of what it is to outright believe that p is to be disposed to assert it in some range of circumstances.³¹ I will remain neutral on which reading is correct because all I need for my argument is that there is this tight connection.

Belief-Assertion Link has a few important features. First, one’s disposition to assert can vary in strength. One’s disposition might be very weak in certain circumstances, e.g., if it’s seriously morally wrong to assert that p in those circumstances. Second, I have left open what range of circumstances one will be disposed to assert in. However, I think that such circumstances are roughly those in which the truth of p is a relevant consideration. For example, one might be defending p against a claim or simply expressing one’s own beliefs or even trying to raise the possibility of p.

When it comes to moral beliefs, one will be disposed to assert that p, where p is a moral proposition in the following kinds of cases: (1) one is giving moral advice in which p is relevant, (2) when one’s audience is ignorant of moral facts that are currently widely accepted because they are older or come from a more isolated community (e.g., think of the times when you’ve had to tell an older family member what the morally permissible way of referring to some group of people is), (3) one is teaching one’s children how to behave (4) one is informing friends and family about unsavory characters or about the immoral behavior that one has witnessed, and (5) one desires the approval or praise of others, and believes one can get it by engaging in moral grandstanding (or virtue signaling).³²

5.3 The Problems with Moral Deference

Given Belief-Assertion Link, I think we can articulate a number of problems with moral deference that together generate a pro tanto reason to avoid it.

First, moral deference leaves the deferrer in the following undesirable position: They outright believe some moral proposition that it would be epistemically and/or morally impermissible for them to assertion and at the same time they are disposed to assert that proposition in some range of cases. Of course, manifestations of this disposition might be defeated if the person realizes that their assertion will be impermissible, but it nonetheless leads to a conflict of motivations. On the one hand, they’ll likely not to want to do something that is epistemically or morally bad, but, on the other hand, they’ll be motivated to do something epistemically or morally bad. I think this gives us good reason to avoid moral deference in general.

Second, there will be cases in which the asserter’s motivation to assert is not defeated because she has forgotten the grounds of her belief.³³ I take it as obvious that we often forget why we believe things—even moral claims. This will especially be true when people form their moral beliefs on, what is for them, an unusual basis, i.e., moral testimony.

Third, people will likely have moral reasons to assert the content of their deferential moral beliefs, e.g., to prevent someone else from acting wrongly or to nudge someone toward doing something supererogatory. Such assertions will often turn out morally permissible and all-things-considered permissible. However, they will involve doing something morally bad, i.e., violating Deal-Breaker, and will be epistemically impermissible. Of course, it’s better that these assertions are not all-things-considered wrong, but they will involve some moral residue when Deal-Breaker is violated. For example, it would be fitting for the asserter to apologize for making the assertion without the proper epistemic credentials and without taking the audience’s epistemic standards seriously. Moreover, these assertions will be epistemically impermissible.

I think that together these problems or the risk of these problems is enough to generate a pro tanto reason against moral deference. Like other pessimists, I think this pro tanto reason will often be outweighed by other considerations. But unlike other pessimists, I think I have shown why moral deference itself puts one in a bad position with regard to both moral and epistemic norms.

Perhaps I’ve overestimated how weak-willed people are concerning asserting moral propositions they epistemically shouldn’t assert. After all, there seem to be many examples of cases in which people seem to have no problem refraining from asserting things they shouldn’t, e.g., doctors tend to maintain doctor-patient confidentiality, lawyers tend to maintain lawyer-client confidentiality, and Catholic priests tend to respect the seal of the confession.

I think this is an important worry. However, I think there are some crucial differences between violating the norm of moral assertion and violating people’s confidence in the above cases. In the above cases, asserting what one shouldn’t is a greater wrong than violating Deal-Breaker, for example. This can be seen by the punishment that doctors, lawyers, and priests can incur for violating people’s confidence. For example, priests can be excommunicated for (either intentionally or unintentionally) violating the seal of the confession, doctors can be fired (e.g., from a hospital) or lose their medical license for violating doctor-patient confidentiality, and lawyers can lose their license to practice law for violating attorney-client privilege. The more serious a wrong is, the more likely good people will be motivated to refrain from violating that wrong—as long as they know it is a wrong. So, we have some reason to think that doctors, lawyers, and priests have reason to not assert certain propositions that moral asserters in violation of the norm of moral assertion do not have.

Moreover, these severe punishments for violating confidence give doctors, lawyers, and priests strong prudential reason to remain silent. But one could violate Deal-Breaker or the norm of moral assertion without the risk of such heavy penalties. This is in part because it’s more difficult to tell if one has violated the latter two norms than it is to tell if someone violated professional obligations involving protecting the confidence of other people.

In fact, one will often have to practical reason to share one’s moral beliefs (even if one cannot epistemically permissibly share them) that doctors, lawyers, and priests don’t have to violate confidentiality agreements. One’s moral beliefs are often accompanied by certain conative states, e.g., desires about how the world ought to be. One reason one has to assert the content of one’s moral beliefs is if one thinks doing so will help make the world more in line with one’s moral desires. In fact, if one thinks that one’s moral beliefs are true, then one should think that one has moral reason to assert the content of those beliefs in order to influence the moral beliefs of other people so that those people behave better.

Finally, as noted above, one might forget the basis of one’s moral belief or mistakenly think that one understands why the moral claim is true or that one can offer an explanation for it. However, it is unlikely that one is going to forget that sharing certain details of another person’s life violates that person’s confidence.

6 An Objection

I’m arguing that connections between moral belief and moral assertion explain what is problematic about moral deference. However, one might be concerned that the explanations I’ve offered cannot explain what is problematic about moral deference involving acceptance instead of belief. In such cases, the deferrer merely accepts the moral claim solely on the basis of a moral expert’s testimony. However, the objection goes, there are no tight connections between accepting that p and asserting that p, so my explanation can’t account for why such moral deference is problematic. Moreover, moral deference via acceptance and moral deference via belief do not seem problematic for different reasons. Rather, there just seems to be one phenomenon that is problematic: moral deference. So, it looks like even if what I say is correct about moral deference via belief, I’m going to have to offer a different explanation of what is wrong with moral deference via acceptance.³⁴

I think there are two ways to respond to this objection. First, there is a partners-in-guilt strategy. In particular, many explanations of what is wrong with moral deference only focus on moral deference via belief.³⁵ Many of these explanations are only applicable to belief.³⁶ So, even if this objection is true, my position is no worse than many others.

Second, I think that a related explanation can explain what is problematic with moral deference via acceptance. In particular, I think it’s plausible that there is an epistemic norm of moral reasoning that parallels the epistemic norm of moral assertion. Part of what it is to accept that p is to treat p as true in one’s practical or moral reasoning and thus accepting that p, where p is a moral proposition, is governed by a norm like the following:

The Understanding Norm of Acceptance: Where one’s choice is p-dependent, it is epistemically permissible to treat the proposition that p as a reason for acting only if one understands (or grasps) why p.

One reason for thinking that this is true is that unless one understands why p is true, one will not be able to weigh the reason that p constitutes against other reasons for and against some action. Understanding why p gives one a grasp of how weighty of a reason p constitutes and without such a grasp, one will not be able to tell if the reason constituted by p is defeated, attenuated, or intensified.

If one is more sympathetic to Simion (2018)’s norm of moral assertion, then one might be more sympathetic to the following norm of acceptance:

The Explanation Proffering Norm of Acceptance: Where one’s choice is p-dependent, it is epistemically permissible to treat the proposition that p as a reason for acting only if one can provide a context-appropriate explanation of why p.

This norm would help explain why some people think that it is morally important to be able to justify your actions to others—especially those directly affected by your actions.³⁷ For example, if you defer that, “It’s morally permissible to hit someone if they annoy you,” then, when you actually hit someone because they annoy you, you’re going to have to justify yourself by explaining why that moral claim is true.

Appealing to the Understanding Norm of Acceptance allows me to offer a unified explanation of the problem with moral deference, because regardless of the kind of deference one engages in, it increases the likelihood that one will violate an epistemic requirement involving understanding or the ability to provide an explanation of why the moral proposition one asserted or acted on is true.

7 Conclusion

In this paper, I argued for a novel version of moderate pessimism about moral deference. I first argued that deferential moral beliefs put one in a bad position to be able to epistemically and morally assert the contents of those beliefs because doing so could involve: (a) violating the norm of moral assertion and/or (b) violating epistemic and moral versions of Deal-Breaker. I then argued that moral deference actually puts one in a worse position than just making it impermissible for one to assert the contents of one’s belief. I argued that there is a tight connection between outright believing some proposition to being disposed to assert it in certain contexts. This puts one in the undesirable position of being disposed, in certain contexts, to violate epistemic and moral norms. Given this, I think moral deference is pro tanto bad from both an epistemic and a moral perspective. Thus, we have pro tanto reason to avoid it.