Against selfless assertions

Abstract

Lackey’s (Noûs 41(4):594–626, 2007) class of “selfless assertions” is controversial in at least two respects: it allows propositions that express Moorean absurdity to be asserted warrantedly, and it challenges the orthodox view that the speaker’s belief is a necessary condition for warranted assertibility. With regard to the former point, I critically examine Lackey’s broadly Gricean treatment of Moorean absurdity and McKinnon’s (Am Philos Q 50(2):121–135, 2013; The norms of assertion: truth, lies, and warrant. Palgrave MacMillan, London, 2015) epistemic approach. With regard to the latter point, I defend the received view by supporting the knowledge account (KA), on which knowledge is the necessary condition for warranted assertion. After examining two defenses of KA, by Montminy (Perspectives in pragmatics, philosophy and psychology, Springer, Berlin, 2013) and Turri (Epistemic norms: new essays on action, belief, and assertion. OUP, Oxford, 2014), I propose two alternative approaches. Although I remain neutral between them, I develop in more detail the view which classifies “selfless assertions” as “presentations”, a type of assertives distinct from genuine assertions. This account is motivated further by allowing for the expansion of the normative approach to other assertives, a feature we may be interested in, in the light of a recent wave of normative accounts of speech acts.

1 Introduction

An assertion that p is selfless if and only if:

• 1. a subject, for purely non-epistemic reasons, does not believe that p;

• 2. despite this lack of belief, the subject is aware that p is very well supported by all of the available evidence; and

• 3. because of this, the subject asserts that p without believing that p. (Lackey 2007: 599)

The class of selfless assertions was originally introduced by Jennifer Lackey (1999) in a debate on the epistemology of testimony (1999: 477–482). In her paper, Lackey employs selfless assertions in an attempt to refute the “necessity thesis” of the transmission principle of testimony, according to which a hearer knows that p on the basis of A’s testimony that p only if A knows that p (1999: 473, 2008: 47–53). Lackey argued that the audience may acquire knowledge on the basis of selfless assertions even though the agent A, by definition, does not know their content.

In her 2007 paper, Lackey evokes selfless assertions for a different purpose: to provide new evidence within the field of normative accounts of assertion. It is with regard to this later project, I believe, that selfless assertions may give rise to suspicion. There are at least two prima facie reasons for this. First, by granting that selfless assertions can be made warrantedly, one rules out the speaker’s belief in p as a necessary condition for warrantedly asserting p—a claim agreed upon by most theorists within the normative tradition.¹ Secondly, by conjoining a report of one’s own attitude towards a selfless assertion, namely disbelief, one thereby utters a Moorean sentence: “p and I don’t believe that p”, and ends up making an assertion which is simultaneously both warranted and absurd.²

These worries might still fail to dissuade. On the one hand, one may pursue a less canonical account of assertion which does not require the speaker’s belief as a necessary condition for a warranted speech act. For instance, Pelling (2013a) and Pritchard (2014) argue that selfless assertions are made warrantedly for reasons pertaining to safety³; McKinnon (2013, 2015) maintains they are warranted as the speakers thereby express knowledge; Wright (2014) appeals to different, virtue-theoretic considerations,⁴ while García-Carpintero (ms1) introduces epistemic obligations.⁵ On the other hand, one may offer an independent explanation of why assertions expressing Moorean absurdity are unwarranted, as does Lackey (2007), by providing a broadly Gricean strategy for dealing with the phenomenon.

In this paper, I defend the knowledge account of assertion (KA): the view that knowledge is the necessary condition for warranted assertibility.⁶ In Sect. 2, I examine two defenses of KA, by Montminy (2013) and Turri (2014), and motivate a departure from such responses. I discuss Lackey’s treatment of Moorean absurdity in Sect. 3, where I claim that her argument rests on the false assumption that Moorean sentences are absurd solely due to being misleading. In the same section, I analyze McKinnon’s approach to Moorean constructions (2013, 2015), which I find more congenial and satisfying. Section 4 develops two alternative takes on selfless assertions, both of which attempt to satisfy the following set of desiderata: (1) they should remain faithful to Lackey’s intuition about the correctness of the speech act performed; (2) they should stay clear of Moorean absurdity; (3) they should preserve the orthodox approach to the necessity of belief for the warranted assertibility; and (4) they should offer an account that obeys the knowledge rule. Although I remain neutral between the two views, I develop the approach which classifies “selfless assertions” as “presentations” in more detail; “presentations” are assertives⁷ distinct from genuine assertions in both the commitment the speaker incurs as well as the characteristic intention one typically has when performing them. The proposal is further motivated by allowing for the expansion of normative approaches to other assertives, a feature we may be interested in, in the light of a recent wave of normative accounts of speech acts.⁸

2 Two responses to Lackey’s argument

To introduce the phenomenon of selfless assertions, let us start by considering the following scenario:

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 (1), though she herself neither believes nor knows this proposition:

• (1) Modern day Homo sapiens evolved from Homo erectus.⁹

Under the assumption that Stella asserts (1) warrantedly—the intuition Lackey’s argument relies on—it is not obvious how accounts that posit norms of belief or knowledge could explain its warranted assertibility.¹⁰ I now turn to two proposals found in the literature¹¹ that defend KA, offered by Montminy (2013) and Turri (2014).¹²

2.1 Montminy

In his brief discussion of selfless assertions, Montminy (2013) accepts Lackey’s contention that Stella, “qua asserter, is appropriately subject to praise” (Lackey 2007: 599). It should be emphasized, however, that Montminy and Lackey understand the condition of being “appropriately subject to praise” qua asserter quite differently.

For Lackey, this condition implies that the relevant assertion is warranted and satisfies the constitutive norm of assertion. In contrast, Montminy denies this, proclaiming Stella’s assertion “epistemically improper” (2013: 47) due to her lack of the relevant knowledge. In Montminy’s view, Stella is praiseworthy only because she chooses to break the linguistic rule in order to obey a more salient, pragmatic rule and thus avoid acting against her professional obligations. “We think she makes the right decision” given that breaking the linguistic rule (i.e., KA) is a “minor” offense compared to violating the pragmatic one (Montminy 2013: 47). Thus, Montminy allows that the assertion is warranted only all things considered, but not epistemically.

This is further confirmed in Montminy’s remark that Lackey’s argument is “like Williamson’s train example” (47). In Williamson’s original scenario (1996, 2009), someone shouts “that is your train” without really knowing that to be the case, concerned that her addressee may miss the train; given the “urgency of the situation” (1996: 508), the knowledge norm is overridden by a pragmatic norm, and the assertion is made warrantedly all things considered. In other words, the fact that one violated the constitutive norm of assertion is secondary, compared to the importance of the practical matters at hand. The same happens, Montminy argues, with the “CREATIONIST TEACHER”. The exact nature of the pragmatic norm may be different from that in Williamson’s train scenario, but it likewise takes precedence over KA.

Montminy’s approach thus suggests that Lackey’s argument from selfless assertions simply restates the old problem of a pragmatic norm outweighing KA. If Montminy is right and Williamson-style considerations do offer a solution,¹³ Lackey’s argument is hardly “the strongest type of counterexample” to KA (Lackey 2007: 598). This is precisely the conclusion that I think ought to be contested. Montminy does not show that selfless assertions are sufficiently similar to Williamson’s train example, but rather circumvents the problem. This is odd, as Lackey is well aware that the proponent of KA can explain the assertion as appropriate “given the circumstances of the situation” (2007: 595) and sets the train example aside as a different type of objection (2007: 595–6).

Another concern with Montminy’s approach is his contention that Stella’s speech act is unwarranted. Montminy thus grounds his defense of KA by denying the intuition Lackey’s argument rests upon. Although I find Lackey’s intuition more natural anyway, I part ways with Montminy’s approach as I am interested in seeing whether KA can explain selfless assertions without having to undermine the foundations of Lackey’s argument.

2.2 Turri

Turri (2014) offers three responses against Lackey’s argument. In his first response, Turri challenges Lackey’s rendering of Stella’s doxastic state¹⁴: contrary to what Lackey concedes, Turri remarks that Stella may actually believe (1); for, if Stella presents (1) as “what is most likely to be true”¹⁵ (Lackey 2007: 599), then, Turri suggests, she mostly believes that (1) is true. Further, Turri takes one’s “mostly believing” that p as sufficient for knowing p. Thus, even if Stella does not believe (1), Stella’s “mostly believing” could still be deemed sufficient for knowledge in this particular case.

Lackey seems to be aware of the possibility that Stella may be seen as “weakly believing” the proposition in question.¹⁶ But in Lackey’s defense, I find it disputable that Stella believes (1) even weakly; we have seen that Stella “regards her duty as a teacher to include presenting material that is best supported by the available evidence” (2007: 599). What this suggests is only that Stella believes that (1) is supported by the scientific data, but not that any degree of credence in (1) per se can be ascribed to her. After all, in all probability, Stella is skeptical of evolutionary biology precisely because it conflicts with Scripture.

According to Turri’s second response, Stella makes a “double assertion”; in the order of explanation, Stella first makes an individual assertion, whereby she speaks only for herself (to be explained shortly) and it is by means of making this statement that Stella produces another, communal assertion, made on behalf of the community of science educators.

Turri then proceeds to demonstrate how KA can explain our ordinary intuitions about the permissibility of both types of assertions. As communal assertion is permissible according to Turri’s view, KA is equipped to account for such an impression of correctness by highlighting that the scientific community, on whose behalf Stella asserts (1), does indeed know the proposition. This much is fairly uncontroversial, I take it. The worry for Turri’s position comes from elaborating on Stella’s individual assertion; to motivate the reader to accept that in such contexts the individual assertions are made impermissibly, Turri considers the following example:

• (2) Speaking just for myself here, modern day Homo sapiens evolved from Homo erectus.

It is fairly easy to agree with Turri that if Stella had asserted the individual assertion (2), then she could not have done so permissibly: due to its added qualifier, we understand that (2) is a lie and thus constitutes an incorrect description of Stella’s assertion. Strangely enough, Turri never shows that the content of (2) does in fact get asserted. He seems to accept that the content of individual and communal assertions Stella makes is the same, viz. that Modern day Homo sapiens evolved from Homo erectus. However, if Stella’s individual assertion does indeed lack a prefixed qualifier, it is not a lie nor is it unwarranted in any discernible way. Thus, although some aspects of Turri’s second response seem to be along the right track, the argument is incomplete. It remains to be shown that Stella’s individual assertion is different from her communal assertion, even though the two assertions are made by uttering one and the same sentence.

Finally, in his third response, Turri tries to support the claim that Stella is not permitted to assert (1), insofar as she does not believe its content, thereby making her assertion unwarranted. In so doing, Turri first appeals to the data from conversational challenges, in particular those centered on the notion of belief. The propriety of asking the speaker: “Why do you believe that?” seems easily explained by the hypothesis that when making assertion, one represents oneself as having the adequate authority, i.e., believing the content. Similarly, the more aggressive replies such as: “You don’t really believe that!” as well as the impropriety of Moorean sentences “p but I don’t believe that p”, seem well handled by the same proposal.

Although I am sympathetic to Turri’s line of thought, his discussion of such challenges is rather short and does not mention recent developments in the literature that defend the Lackeyan approach.¹⁷ What is more, Turri seems to overlook Lackey’s own attempt to meet the objection concerning Moorean sentences. I now turn to examine Lackey’s view and extend Turri’s point.¹⁸

3 Can we warrantedly assert Moorean constructions?

Arguably the most pressing worry for an advocate of selfless assertions comes from Moorean sentences: “p but I don’t believe that p”.¹⁹ If the first conjunct is a selfless assertion, it seems that whenever we deem p warranted and the second conjunct a sincere assertion, the whole conjunction should be evaluated as warranted, too:

• (3) Modern day Homo sapiens evolved from Homo erectus but I don’t believe it.

In what follows, I discuss two treatments of Moorean absurdity offered by Lackey (2007) and McKinnon (2013, 2015).

3.1 Lackey

According to Lackey’s Reasonable to Believe Norm of Assertion (RTBNA), an assertion that p is warranted only if: (1) it is reasonable for one to believe that p; and (2) if one asserted that p, one would assert that p at least in part because it is reasonable for one to believe that p (Lackey 2007: 608). Selfless assertions satisfy both clauses of the RTBNA and are rendered warranted.²⁰ The result is problematic, as Moorean sentences are often evaluated as absurd, extremely odd (Levin 2008: 376), or even defective (Schaffer 2008: 8). Recognizing the difficulty Moorean sentences pose for her view, Lackey proposes the following strategy.

In developing her response, Lackey introduces the following scenario, where an assertion is misleading although not absurd in the Moorean sense:

LOSING DRINKER: Nadia and Hank know both that their friend Nina tends to go to the bar only when she loses a tennis match and that this is a fact that is generally known by all of her friends. However, Nadia knows further that Nina went to the bar today to have a drink with her opponent despite having won her tennis match. Nevertheless, while discussing Nina’s recent tennis matches, Nadia asserts to Hank, “Nina went to the bar earlier today after her tennis match” (Lackey 2007: 614).²¹

Lackey contends that no normative account of assertion is equipped to assess Nadia’s assertion as unwarranted. For this reason, she insists, all normative accounts should adduce the Not Misleading Norm of Assertion** as “an additional norm governing assertion” (2007: 615):

NMNA**::

S should assert that p in context C only if it is not reasonable for S to believe that the assertion that p will be misleading in C relative to the purposes of the exchange in question.²²

According to this view, both Stella and Nadia would mislead their audience by making their respective assertions, as their audience would either form false beliefs or no relevant beliefs at all (Lackey 2007: 615).

Now, Lackey maintains that an assertion is absurd in the Moorean sense due to being misleading. As applied to “CREATIONIST TEACHER”, Stella would be offering “either too little or too much information” when asserting: too much for adding the second conjunct in the first place, or too little if they did not explain why it was added (Lackey 2007: 616). I take this to fall short of clarifying what is truly problematic about Moorean sentences. After all, Lackey mentions that selfless assertions are merely “most likely” to lead to such consequences (615), leaving open the possibility that the audience may not be misled.

Given that the RTBNA does not have the resources to deem selfless assertions unwarranted, Lackey proceeds to use the NMNA** for this purpose. Accordingly, Lackey describes Stella’s assertion as “improper” (617) and “impermissible” (619). The use of these adjectives strongly suggests that the NMNA** is added to the RTBNA, rather than merely being a pragmatic norm, as the text might suggest.²³

That said, one may wonder if the addition of the NMNA** is actually needed. First, Lackey does not provide a detailed account of the relation between an assertion being misleading and being unwarranted. Secondly, the NMNA** is more aptly viewed as an instance of a general rule of conversational exchange, which can be rephrased for any illocutionary force. (One may substitute “assert” in the NMNA** for “ask” or “guess” and obtain equally good rules.) As applied to “LOSING DRINKER”, Nadia’s fault lies only in her uncooperative behavior: her assertion seems warranted as she could not be epistemically positioned any better than she is. Thirdly, the proponent of a reasons-based account could maintain that Stella’s assertion of (3) is problematic insofar as it generates a misleading implicature, and yet argue that it is warrantedly assertible. Similar accounts have in fact been proposed in the literature.²⁴

In short, Lackey’s treatment of Moorean absurdity as a pragmatic phenomenon is unsuccessful. The explanation that whenever a Moorean sentence is uttered, the ensuing act is absurd because it is misleading to the hearers and there is nothing further “absurd” about it, misses the point. The Moorean sentence (3) cannot be treated analogously to “LOSING DRINKER”, as the differences between the two cases are striking.

3.2 McKinnon

McKinnon’s (2013, 2015) offers an alternative treatment of Moorean absurdity. Unlike Lackey, McKinnon insists that selfless assertions can be warranted simpliciter.²⁵ In order to examine McKinnon’s approach, it will be useful to outline her general account of assertion.

McKinnon argues that an assertion that p is warranted only if the speaker has “supportive reasons” for p,²⁶ i.e., “particularly epistemic support for the assertion that p” (2013: 53). To flesh out the notion of supportive reasons in more detail, let us revisit Stella. When asserting (1), “Modern day Homo sapiens evolved from Homo erectus”, Stella has “the aim of having her students come to believe what is best supported by the available evidence” (70). The fact that Stella has such an “epistemic goal” (71) provides supportive reasons for her assertion, rendering it correct. In contrast to Stella, we can imagine Norbert who does not believe that there is any evidence in favor of the theory of evolution (71). As a creationist biology teacher, Norbert asserts (1) only because his professional duties dictate this course of action to him. Consequently, McKinnon urges, Norbert asserts (1) unwarrantedly, as his reasons fall short of being supportive in nature.²⁷

The key to understanding McKinnon’s analysis of Moorean absurdity is that Stella’s supportive reason for offering (3) is the same supportive reason she had for asserting its first conjunct alone (1).²⁸ In both cases, Stella is “properly aiming” at the intended goal insofar as she is expressing knowledge. It is precisely this property, according to McKinnon, which makes Stella’s selfless assertion warranted.

The appealing feature of McKinnon’s approach to Moorean assertions is that, although they appear absurd, they can be warranted as long as the speaker is properly aiming at expressing knowledge (McKinnon 2015). In comparison to Lackey’s account, I find this view more promising.²⁹ In what follows, I will likewise offer an account which treats Moorean constructions as “appearing” to be absurd, while insisting they are in fact warranted. That said, let me express one reservation one may have regarding McKinnon’s account.

McKinnon allows that one can warrantedly assert a proposition one knows to be false. In her Jenny scenario (2015: 61–72; 204–6), McKinnon describes a physics teacher who falsely asserts that electrons behave according to the Bohr model because she knows her students would not understand the received, valence model, and thus offers the Bohr model as a “stepping stone”. While we can accept that Jenny has supportive reasons, we may doubt with Goldberg (2015b) that this type of intuition is founded: Jenny’s supportive reasons might be better viewed as her following the practical norm which outweighs the linguistic one, and her assertion being warranted all things considered. Indeed, it is somewhat counterintuitive to insist that both electrons behave according to the Bohr model and electrons do not behave according to the Bohr model can be viewed as satisfying the constitutive rule of assertion.

I am not suggesting that these worries cannot be overcome, however. Rather, I wish to offer a strategy that may avoid such complications and thus have a comparative advantage over McKinnon’s view. I will argue that when Stella utters (1), she asserts that according to the theory of evolution, (1) is true. Accordingly, when Stella utters (3), her assertion is not absurd in the Moorean sense because what she disbelieves is that man evolved from apes, but not that this proposition is true according to Darwin’s theory. The same follows for Norbert as he likewise believes that (1) is true according to the theory of evolution, in virtue of being a biology teacher. In what follows, I discuss this strategy in more detail.

4 In defense of the knowledge account

In the remainder of the paper, I outline two approaches to selfless assertions along the lines of KA. Both responses retain the orthodox belief condition concerning the correctness of assertion, and neither response presents the speaker as committed to Moorean absurdity. In addition, they both preserve Lackey’s intuition that the illocutionary act performed is warranted.

The two approaches attempt to explain how one gets into the situation where “selfless assertions” occur and are made warrantedly. We should observe that “CREATIONIST TEACHER” features utterances made by “spokesperson” of a certain profession. As far as the teaching profession is concerned, any teacher uttering (1), whether they share Stella’s convictions or not, seems required to defend only the following, hedged content:

• (4) According to the best available evidence, modern day Homo sapiens evolved from Homo erectus.

To see this, note that if a teacher were challenged for uttering (1), it would not be necessary to defend evolutionary theory, but only (4). What is more, it would not be pragmatically odd for any teacher to add that personally they prefer an alternative to (1). Finally, any attempt to press teachers to defend the stronger (1), rather than (4), would be out of place.

Certain contextual parameters pertaining to the act of teaching, we may say, contribute to one conveying weaker content than we may think simply by looking at the sentences uttered. Yet, once a teacher is out of the classroom, those contextual parameters are no longer in force, and uttering (1) would then commit that person to defending³⁰ precisely (1), and not (4).

To explain these remarks, the first proposal contends that what is asserted is only the hedged content: (4). Prefixing the propositional content with the evidential³¹ “according to the best scientific evidence” does double duty insofar as it explains that Stella’s assertion is neither selfless nor unwarranted: Stella firstly believes the content of (4) and secondly knows it.

Note that if such “contextual parameters” pertained with respect to all assertions, the consequences would be counterintuitive: we would not be able to make flat-out assertions, but instead only guardedly express the content prefixed by the appropriate evidential (“I have read that p”, “I was told that p”, and so on). So what is it that prevents the proposed strategy from overgeneralizing?

First, notice that such contextual parameters can be cancelled within professional settings themselves; if Stella prefixed (1) by “in my opinion”, she would assert precisely (1), not (4), despite addressing the audience as a teacher. The “in my opinion” would signal she is now committing herself to defending (1).

Secondly, assume by reductio that by uttering “it is 6 o’clock” one asserts the hedged content: “according to my watch, it is 6 o’clock”. If we assume further that one’s watch stopped working and it is actually 8 o’clock, the intuition we have is that the assertion would nonetheless be unwarranted. However, if one indeed managed to assert the hedged content, “according to my watch, it is 6 o’clock”, the assertion should be warranted instead: it is both true and known by the speaker. Thus, in deciding whether one asserts the straightforward p or a hedged “evidential, p”, a natural suggestion is to ask ourselves whether the speaker is committed to defending the former or only the latter content.

At this point, one may object that the explanation of why Stella is committed to defending only (4) stems from her profession. Accordingly, Stella is committed to defending (4) not because she asserts (4) but because it is her professional obligation to defend only (4) and not (1).

In reply, I find the remarks on the origin of Stella’s commitment misguided. If Stella is indeed committed to defending (4), i.e., if she licenses others to act on (4), if it is true that when challenged she must defend (4), and so on, then she indeed asserts (4). By clarifying the origin of Stella’s commitment one does not show that (4) doesn’t get asserted; instead, one only fleshes out the general idea of “contextual parameters” referred to above.

4.1 Creationist teacher

The second proposal attempts to account for the same intuition: that Stella is not committed to defending (1) when she utters (1) in the classroom. The explanation is now different, however, as according to this view, Stella is not making an assertion to begin with; instead she is performing a different illocutionary act. It is by focusing on the nature of this illocutionary act that we can explain why teachers are committed only to defending the weaker (4) in virtue of uttering (1).

In introducing “CREATIONIST TEACHER”, Lackey describes Stella as “presenting material that is best supported by the available evidence” (599, italics added).³² According to my second proposal, “presenting” denotes an illocutionary act Stella performs. Although it is an assertive, the speaker’s obligation when presenting given content is different from that associated with other assertives.³³ For, by presenting p, one commits oneself to defending that p is true according to the given source (in Stella’s case, according to the scientific theory she presents), and not to p simpliciter ³⁴ as is the case with assertoric commitments.³⁵

To motivate this strategy, note that it is a mirror image of the first strategy: while in the first approach I argued that, as opposed to Lackey’s view, the content changes but the illocutionary force remains the same, I now explore the possibility that the content remains the same, but that the force is different. Further, in support of the second approach, observe that Stella merely teaches a theory [or a part of it, such as (1)] and therefore can deny that she is thereby strictly speaking asserting it. As already mentioned above, if Stella was challenged for uttering (1), she could reply by defending (4) as true without making the case for (1). That said, since Stella must justify (4) and not (1), we may say that the illocutionary act teachers perform is one by which, in uttering a sentence s, one commits oneself to defending that s is true according to the theory one is presenting.³⁶

Furthermore, as Pelling (2014) argues, if two illocutionary acts are different, then the intentions characteristically associated with them will probably be different too. This seems to be corroborated in the case of presenting p and asserting p: the speaker’s communicative intention while presenting p (say, as a piece of scientific theory) is to undertake responsibility for the truth of p relative to such a scientific view, but not to defend p itself. The latter intention, to defend p, is characteristic of assertion, and in the present case is something we would expect an evolutionary biologist to have.

Here, the worry concerning overgeneralization reappears: if we claim that Stella is performing a different illocutionary act, one may wonder why our ordinary assertions should not also be viewed as presentations, that are made relative to the relevant sources. By means of example, it might be that by uttering “it is six o’clock” I am merely presenting that it is six o’clock according to my watch, but not straightforwardly asserting that it is six o’clock.

My reply here parallels the one given before: to settle on which illocutionary act is being made, we need to understand which content the speaker is committed to defending; even if my watch is broken, my act will be normatively incorrect (i.e., I will have to retract it, once I realize its content is false). This suggests that the content conveyed is not that associated with presenting; for, as argued above, one presents that p always relative to some source. Thus, if this were a case of presenting, our normative intuitions should be precisely the opposite from those they are. In contrast, even if (1) were false, Stella’s speech act would still appear normatively correct, which might be due to the fact she only presents (1) relative to a given theory.

In short, while we assert (1) warrantedly only if we know that (1), we present (1) warrantedly if we know that (4). By endorsing either of the two views proposed, we avoid the accusation that Stella asserts warrantedly what she does not believe. In keeping with the first proposal, Stella asserts (4) and does so warrantedly because she knows (4). In keeping with the second, Stella presents (1) and does so warrantedly because she knows that (1) is true according to the best available evidence. This is a normative account of presenting that we may find plausible, provided we previously accept KA. Furthermore, in keeping with either of the two approaches, Stella is not committed to Moorean absurdity (5) but only to an uncontroversial (6):

• (5) Modern day Homo sapiens evolved from Homo erectus but I don’t believe that.

• (6) According to accepted scientific consensus, modern day Homo sapiens evolved from Homo erectus but I don’t believe that Homo sapiens evolved from Homo erectus.

4.2 Distraught doctor

In the remainder of the paper, I apply the second strategy to two further scenarios: “DISTRAUGHT DOCTOR” and “RACIST JUROR”.

DISTRAUGHT DOCTOR. Sebastian is an extremely well-respected pediatrician and researcher who has done extensive work studying childhood vaccines. He recognizes and appreciates that all of the scientific evidence shows that there is absolutely no connection between vaccines and autism. However, shortly after his apparently normal 18-month-old daughter received one of her vaccines she was soon diagnosed with autism. While Sebastian is aware that signs of autism typically emerge around this age, regardless of whether a child received any vaccines, the grief and exhaustion brought on by his daughter’s recent diagnosis cause him to abandon his previously deeply-held beliefs regarding vaccines. Today, while performing a well-baby checkup on one of his patients, the child’s parents ask him about the legitimacy of the rumors surrounding vaccines and autism. Recognizing both that the current doubt he has towards vaccines was probably brought about through the emotional trauma of dealing with his daughter’s condition and that he has an obligation to his patients to present what is most likely to be true, Sebastian asserts, “There is no connection between vaccines and autism.” In spite of this, at the time of this assertion, it would not be correct to say that Sebastian himself believes or knows this proposition. (Lackey 2007: 598-9)

To take a slightly different starting point, observe that while Sebastian’s uttering of (7) would not strike us as redundant, his uttering of (8) in the same circumstances would:

• (7) There is no connection between vaccines and autism.

• (8) According to the best available evidence, there is no connection between vaccines and autism.

Again, this is because, in asking for a professional opinion, we are interested in hearing precisely what the best available evidence suggests. Thus, analogously to the case of Stella, we can say that in presenting (7), Sebastian commits himself to defending (8), without asserting either of the two; like Stella, Sebastian is a “spokesperson” for his profession and has “an obligation to his patients to present what is most likely to be true” (2008: 599). In presenting (7), Sebastian is doing what every pediatrician would be doing in his place. Relatedly, Sebastian does not mention the rumors, nor express doubts motivated by what happened to his daughter. Finally, as in the case of Stella, Sebastian does not warrantedly assert what he disbelieves nor does he commit himself to Moorean absurdity of any kind.

4.3 Racist juror

Lastly, let us work through “RACIST JUROR”.

RACIST JUROR. Martin was raised by racist parents in a very small-minded community and, for most of his life, he shared the majority of beliefs held by his friends and family members. After graduating from high school, he started taking classes at a local community college and soon began recognizing some of the causes of, and consequences of, racism. During this time, Martin was called to serve on the jury of a case involving a black man on trial for raping a white woman. After hearing the relatively flimsy evidence presented by the prosecution and the strong exculpatory evidence offered by the defense, Martin is able to recognize that the evidence clearly does not support the conclusion that the defendant committed the crime of which he is accused. In spite of this, however, he can’t shake the feeling that the man on trial is guilty of raping the woman in question. Upon further reflection, Martin begins to suspect that such a feeling is grounded in the racism that he still harbors, and so he concludes that even if he can’t quite come to believe that the defendant is innocent himself, he nonetheless has an obligation to present the case to others this way. Shortly after leaving the courthouse, Martin bumps into a childhood friend who asks him whether the “guy did it.” Despite the fact that he does not believe, and hence does not know, that the defendant in question is innocent, Martin asserts:

• (9) No, the guy did not rape her. (Lackey 2007: 598)

Start by noticing that Martin utters (9) twice: once in a “high stakes” context (to his fellow jurors in the courthouse), and once in a “low stakes” context (to his friend over a lunch).³⁷ While in the courthouse, Martin has an obligation to present the case according to what the evidence seems to suggest. This kind of obligation is common to all three scenarios, and so Martin’s first utterance of (9) can be understood analogously to the two cases previously discussed—as a speech act of presenting. More precisely, since Martin must deliberate relative to the standard set of legal rules, which do not allow for a biased conclusion, by uttering s Martin commits himself to knowing the content of: “according to the evidence presented at court, s”. Thus, in presenting (9), Martin commits himself only to the content of (10):

• (10) No, according to the evidence presented at the court, the guy did not rape her.

The reason Lackey introduces “RACIST JUROR” is to avoid tying selfless assertions to professional settings; thus, unlike Sebastian and Stella who made statements at their places of work, Martin speaks outside the courthouse and hence lacks the obligations he would otherwise have as a juror (2007: 601). This should make us rethink our strategy; even if we grant that Martin did not make any assertion in the court, it seems that he must be doing so now (and moreover doing it selflessly). To support this worry, note that by adding the prefix to Martin’s original assertion, as in (10), the resulting assertion Martin addresses to his friend would not be redundant as it was in the case of Sebastian.

A point worth noting is that Martin repeats (9) during the lunch even though he is now not under any obligation to do so. As far as I can tell, (9) is either a lie or it is not a selfless assertion. To show this, note that in these particular circumstances, both (10) and (11) seem to be better candidates:

• (11) Yes, the guy did it.

Lackey could still object that while (10) may confuse the hearer as to why Martin is not offering his own view on the matter, (11) may require further justification which Martin might not be prepared to offer. All this notwithstanding, uttering (9) in the given low-stakes context should count as lying. Martin asserts what he does not believe while there are no pragmatic considerations compelling him to do so. Lackey, however, disagrees that Martin is lying:

[T]here is absolutely no intention on the part of the asserter to deceive or otherwise mislead. Indeed, quite the contrary is true—the asserter in question positively intends to not deceive or mislead her hearer and, as a result, asserts what she herself does not believe. (Lackey 2007: 602)

In support of her claim, Lackey calls attention to Augustine’s two necessary conditions for lying: the speaker must (a) assert that p without believing it and (b) assert that p with the deliberate intention of deceiving.

This may be too quick, however. For one thing, (b) is arguably not a necessary condition for lying: as Carson (2006), Sorensen (2007) and Fallis (2009) have argued, bald-faced lies seem to pose a counter-example to Augustine’s conception. Regardless of that, (b) in fact holds in Martin’s case: although Martin does not mislead his friend about the decision made at court, he is still deceiving his hearer about his own belief. Moreover, it is plausible that it is strictly the content of Martin’s belief that his friend is interested in. Indeed, by repeating the judge’s verdict, Martin would not count as answering “whether the guy did it” (as required by the scenario), but only what the final verdict was, what the judge declared concerning this issue. Hence, if Martin were really asked to give his own take on the issue, he would be lying by offering (9).

Assume, for the sake of the argument, that (9) is not a lie. In that case, it is most plausible to think that Martin chose to utter (9) to hide his racist views. In so doing, Martin would play along as if he was being asked about the court decision. But in this case, Martin would be saying what he believes and hence would not be making a selfless assertion.

Finally, a potential concern may be that in uttering (11), Martin would be covering up the conclusion made by the court. To prevent this, it is more pragmatically appropriate to offer a fuller reply:

• (12) According to the evidence presented at the court, the guy did not rape her. But if you’re asking me what I think, I still think he did it.

As Martin believes both conjuncts, no selfless assertion is made by uttering (12), either.

5 Conclusion

I have offered two alternative accounts of selfless assertions which would satisfy the following set of desiderata: (1) they should remain faithful to Lackey’s intuition that the speech act performed is warranted; (2) they should stay clear of Moorean absurdity; (3) they should preserve the orthodox approach concerning the necessity of belief for warranted assertibility; and (4) they should offer an account that obeys the knowledge rule. While I try to stay neutral between these two accounts, I end up expanding on the version according to which selfless assertions are viewed as a type of assertive distinct from assertion, namely presenting. In so doing, I argue that performing these acts commits the speaker to knowing the propositional content prefixed by a relevant evidential. Such an account is motivated by allowing for the expansion of the normative approach to other assertives, a feature of interest, in light of the recent wave of normative accounts of speech acts.