Dissonance and Moorean Propositions

Dissonance and Moorean Propositions Cristina Borgoni (Penultimate draft. Final version to appear in Dialectica) ABSTRACT In a dissonance case, a person sincerely and with conviction asserts that P, while her overall automatic behavior suggests that she believes that not-P. In contrast with several mainstream views, this paper defends the contradictory-belief view of some relevant dissonance cases and explores its consequences regarding Moorean propositions. The paper argues that in relevant cases, the dissonant person is justified in asserting (or believing) a Moorean proposition on the grounds of her explicit view on the subject matter and the recognition of her opposing beliefs. The person is irrational in being dissonant, but not in asserting a Moorean proposition. Keywords: dissonance, irrationality, Moorean propositions, self-knowledge, belief selfascription Introduction Something intriguing happens when one’s sincere assertions do not match one’s own behavior. Suppose that someone sincerely and with conviction makes an assertion that clearly conflicts with her overall behavior. An appealing diagnosis is that this person does not believe what she herself is asserting. I argue that, in some cases, a better diagnosis is that this person has two mutually contradictory beliefs. If I am right, such a person is justified in 1 asserting (or believing) a Moorean proposition1 of the type ‘P but I believe that not-P’, given her explicit view on the subject matter and the recognition of her opposing beliefs. In such cases, a Moorean sentence, a well-known problematic sentence type, is both justified and accurate.2 This paper has two interrelated objectives. One is the defense of the contradictorybelief view of some relevant dissonance cases. 3 Despite being a possible and plausible reading of such cases, the view has received little attention in the literature. The second objective is to explore the consequences of grasping one’s own dissonance in terms of contradictory beliefs. Despite the increasing interest in dissonance cases, the perspective of the dissonant person on her own dissonance is still in need of closer treatment. In this paper, I argue that once the person grasps her own dissonance, she is in a position to assert (or believe) a Moorean proposition. Some competing readings of dissonance cases seem to drive similar results regarding the availability of Moorean propositions to the dissonant person. However, the one defended in this paper has a further advantage. It can accommodate the irrationality of those in the position to assert or believe Moorean propositions. This paper’s view is that the 1 A Moorean proposition is a proposition that can be expressed by one of the two paradoxical Moorean sentences ‘P but I do not believe that P’ and ‘P but I believe that not-P’. These sentences are often called ommisive and commissive Moorean sentences, respectively. My use of ‘proposition’ is not committed to any particular theory of propositions. Given the class of dissonance cases discussed in this paper, I focus on commisive Moorean propositions. 2 The idea that there are legitimate Moorean assertions is not completely new in the literature (see Turri 2010; Pruss 2011). However, there is no detailed analysis of such assertions in the context of a dissonance case. Chislenko (2014) has recently argued that there are legitimate Moorean and belief-akratic-paradoxical beliefs when defending the possibility of akratic beliefs. However, Chislenko does not enter into the discussion of what mental states underlie cognitive dissonance. Furthermore, although there might be an overlap between epistemic akrasia and dissonance cases, they do not single out the same set of cases. A dissonant individual does not need to hold an akratic belief (Borgoni 2014b), nor does the akratic individual need to be dissonant in the sense discussed in this paper. 3 This paper defends the contradictory-belief view within a pluralist view of dissonance cases. Different psychological phenomena and different combinations of mental states can give rise to similar patterns of dissonance between the individual’s sincere assertions and automatic attitudes. So, a unified interpretation of all dissonance cases is likely inaccurate. I will return to this point in section 3. However, the focus of this paper is the defense of the contradictory-belief view as the best interpretation of key dissonance cases, such as the one proposed in section 1. 2 person is irrational in being dissonant, although not irrational in asserting (or believing) a Moorean proposition. Being irrational in the particular way that gives rise to a dissonance case is what puts the subject in the position to assert and believe a reasonable and justified Moorean proposition. The paper proceeds as follows. Section 1 lays out the paper’s working case. It develops a provisional argument in which the dissonant person is in a position to assert (or believe) a Moorean proposition. Section 2 addresses the question of the subject’s psychology underlying a dissonance case. It defends the contradictory-belief view over the main competing readings of the case. Section 3 responds to possible criticisms of this paper’s position. Section 4 briefly discusses some views on Moore’s paradox. It suggests that a dissonant person might be irrational in being dissonant, but not in asserting a Moorean proposition. Section 5 revises the argument of Section 1. A specific form of Moorean assertion (and belief) can be justified on the grounds of the evidence on a given subject matter and the evidence regarding one’s own beliefs. 1. The Assertion-Behavior Dissonance Case Consider the following case: EMILIA:4 Emilia is an academic who sincerely and with conviction asserts that men and women are equally competent in politics. Her assertion is the conclusion of her reasoning from considerable evidence for that claim. Emilia has examined in detail the question of whether different genders have equal capabilities for political activity. She is prepared to argue for their equality. In fact, she is actively engaged in discussions about 4 EMILIA (Borgoni 2014b) is inspired by Schwitzgebel’s (2010, 2011) cases, ‘Juliet, the implicit racist’ and ‘Ralph, the sexist’. 3 sexism. However, Emilia is systematically sexist in most of her automatic behavior. For example, she is biased when listening to female politicians’ speeches. She is uncomfortable and insecure about women being her representatives. She is also biased when evaluating candidates for the next election. To her, women’s proposals rarely seem as good as men’s. When they do, she is surprised by this fact. Nevertheless, self-deception is absent in many respects. Emilia is aware of her biases and even tries to control them but, of course, she cannot monitor her biases at all times. EMILIA exemplifies an ‘assertion-behavior dissonance case’ or just a ‘dissonance case’. This is a case in which a person sincerely and with conviction asserts that P while her overall automatic reactions suggests that she believes that not-P, despite being aware of the dissonance. Let ‘P’ be the content of Emilia’s assertion. Emilia comes to assert that P by reasoning from the evidence about P. Emilia is consciously committed to the truth of P when she asserts that P. Her explicit assertions contrast with her overall automatic reactions. Such reactions include automatic, instinctive, and unguarded linguistic and non-linguistic behavior. They also include some phenomenological responses. I call this group of reactions ‘abehavior’. Emilia a-behaves as if she believes not-P. As the case is described, Emilia is aware of her dissonance and seems to be in a position to know she is dissonant. In order to understand how Emilia accesses her dissonance, let us rely on the intuitive difference between the first- and third-person perspectives on oneself. These perspectives are associated with the two generic ways by which we can know our psychological states. The first-person perspective on oneself is associated with the special way by which we ascribe mental states to ourselves. It is a common assumption that such a perspective has special features that explain how we can know some of our mental states in a privileged 4 way.5 It differs from the way others ascribe mental states to us. In contrast, the third-person perspective on oneself is associated with the way other people come to attribute mental states to us. It is also called ‘a theoretical perspective on oneself’ (Moran 2001). Such a perspective works analogously to any third-person perspective. From such a perspective, we use evidence that any person could use to learn about another person’s mind. Such evidence includes, for example, one’s behavior and the states of one’s body. It also includes evidence from others’ testimony about one’s behavior and mental states. Emilia can know what she assents to from the first-person perspective. She knows, in a privileged way, that she is committed to the truth of P. She could know it from the third-person perspective as well. However, she does not need to hear herself asserting that P to know that she assents to P. In contrast, she does need to rely on evidence about herself to figure out what mental state underlies her a-behavior. She also needs to rely on the evidence of her mind to figure out that her a-behavior contrasts with her expressed convictions. For this paper’s purposes, I will assume that Emilia carries out reasoning from the third-person perspective to discover her dissonance.6 Now suppose that Emilia not only a-behaves as if she believed that men and women are not equally competent in politics, but she in fact believes such a thing. Under this assumption, Emilia could assert that P but believe that not-P. Taking the third-person perspective on herself allows Emilia to discover whatever we can discover about her mental 5 Different philosophical theories have focused on different features, such as: first-person authority (see Wright 1998, Davidson 1984), immunity to error through misidentification (see Shoemaker 1968, Evans 1982) and transparency (see Evans 1982, Moran 2001). This paper is not committed to any particular explanation of such special epistemic features. It only assumes that there is a special and epistemically privileged way by which we know our own thoughts. 6 Whether acquiring self-knowledge based on the evidence of one’s beliefs could count as a particular type of first-personal knowledge is an open question. Emilia, for example, has access not only to external evidence about her sexist belief, but also to the phenomenology linked to that belief. For present purposes, answering the previous question is not central. The paper only assumes that the way Emilia knows her sincere assertions is different from the way she knows what underlies her a-behavior. However, this view is compatible with the idea that both ways are first-personal. 5 states. Thus, the previous reading of her dissonance is also available to Emilia. She can reasonably affirm ‘I assert that P but I believe that not-P’. In asserting that P, Emilia assents to P. She is in a position to say (and believe), ‘P but I believe that not-P’. Emilia’s situation is such that she can assert a reasonable and justified Moorean proposition. In what follows, I will defend one particular reading of the case that supports the above supposition, according to which Emilia believes that not-P. I will defend the view that Emilia has two mutually contradictory beliefs. Given such a reading, Emilia will be in a position to assert (and believe) a Moorean proposition. 2. The Psychological Dissonance Underlying a Dissonance Case This section explores two mainstream accounts of dissonance cases and defends the contradictory-belief view of cases like EMILIA. Let us begin by examining the account derived from Gertler’s work (2011): 2.1 Emilia judges that P, but lacks a correspondent belief that P. She believes that not-P. In her criticism of transparency accounts of self-knowledge, Brie Gertler (2011) seems to suggest an account of dissonance cases (cf. Schwitzgebel 2010) that supports (2.1). 7 Such an account relies on the difference between judgments and beliefs. According to Gertler (2011: 130), a judgment (an occurrent belief, in her terms) is an act that can easily come and go. As an act, it is a ‘tiny fraction’ of one’s mental configuration. In contrast, a belief (a dispositional state, according to her) is more stable. Since a belief is a state and a judgment is an act, they differ with respect to their stability in the face of new evidence. When one 7 It is worth emphasizing that (2.1) is only derived from Gertler’s work since she does not defend (2.1) in those terms. Gertler concedes the possibility of interpreting the case in terms of the contradictory-belief view, but her interpretation of a similar case to EMILIA favors a reading like (2.1). Cassam (2010) and Peacocke (1998) also seem to favor (2.1). 6 judges, one can reason correctly—or incorrectly—from one’s evidence even though one’s preexisting beliefs can remain the same. Gertler’s discussed example shares some traits with EMILIA. In her example, Nick is the dissonant person whose explicit views contrast with his a-behavior. Nick was inculcated in childhood with the belief that spilling salt brings bad luck. However, today, he repeatedly judges such a belief to be pure superstition. He has enough evidence to judge that spilling salt does not have anything to do with luck. Nevertheless, he cannot avoid feeling an immediate sense of doom whenever he spills salt. According to the suggested reading, Nick’s a-behavior is the expression of his belief while his judgment about his superstition is a mere judgment, and as such, it represents only a tiny fraction of his mental configuration. Applying this approach to EMILIA, Emilia’s assertion manifests her judgment, and not a belief, that P. Her judgment is sensitive to the evidence supporting P. In contrast, Emilia’s a-behavior manifests her belief that not-P, which resists the evidence that she possesses. If we accept that Nick and Emilia’s a-behaviors express their beliefs, such beliefs are clearly disconnected from their explicit judgments. According to some authors, such cases exemplify when it is possible to judge that P and yet fail to believe that P (Cassam 2010: 8182). 8 However, it is not obvious that dissonance cases like EMILIA and Nick’s exemplify situations in which there is a judgment without a corresponding belief. For the subject might hold two contradictory beliefs. Especially in cases like EMILIA, the judgment does not represent a mere isolated sliver of the subject’s psychology. Emilia has a pattern of behavior also linked to that judgment (i.e. she defends equalitarianism, she has researched the subject matter, etc.). It seems reasonable to accept that in EMILIA, there is a belief that guides 8 To be fair to Cassam (2010), he holds that in the normal case, a judgment enters into the formation of a belief. 7 Emilia’s a-behavior, which is not related to her judgment. But her having such a belief does not preclude Emilia from having another belief formed from her judgment. So position (2.1) as an interpretation of cases like EMILIA is not obviously correct. Indeed, there are good reasons to think that it is actually incorrect. It is reasonable to posit an ontological distinction between belief and judgment. And there might be some cases in which an individual judges something without the judgment having any impact on his or her psychological configuration. This could happen in a case in which the individual judges that P and immediately forgets his or her conclusion. However, this is not the case of Emilia or Nick. Their explicit and repeated judgments are linked to other mental states (i.e. they believe the reasons in support of their judgments, and they hold other beliefs inferentially connected to those judgments). They do not forget what they have previously judged. Furthermore, their judgments are also connected to a broad set of behavior. Interpreting Emilia’s explicit assertion as an expression of a belief state seems to be the most reasonable interpretation of what the mental correlate of her assertion is. In EMILIA, Emilia comes to the assertion that P by reasoning from the evidence. She is committed to the truth of the equality of the genders in respect to their political abilities, and this is to believe such a thing. If a judgment is a conscious commitment to the truth of a proposition that one arrives at by reasoning from the evidence, there is not much space to separate it from believing the judged proposition. And in her case, the act of judgment is part of the formation of such a commitment. It does not follow that the formation of every belief involves a judgment.9 Emilia’s supposedly opposing belief, which guides her a-behavior, could have been formed without her making any judgment at all (such as Nick’s belief that was inculcated in him from childhood). Although the judgment itself 9 A belief is not necessarily linked to, nor formed from, a judgment (Cassam 2010: 83). We acquire beliefs in several different ways, including ways that are not governed by rational patterns. 8 can be considered a tiny slice of her mental configuration, the formation of such a judgment has a broader impact on her mind; it forms a correspondent belief. So far, I have been arguing that Emilia’s assertion also counts as manifesting a belief. Such an argument does not deny that Emilia’s a-behavior manifests another belief of hers. Emilia has two mutually contradictory beliefs. These beliefs somehow operate very differently from one another. One is evidence-oriented, while apparently the other is not. The latter is resistant to the evidence Emilia has rationally collected. However, in order to further defend the contradictory-belief view of this class of cases, it is necessary to respond to another competing reading, such as the one proposed by Tamar Gendler. According to Gendler’s line of reasoning, if Emilia’s beliefs operate so differently, we should not consider them as psychological states of the same type. Gendler’s (2008) interpretation of dissonance cases employs the notion of aliefs. Based on her account, EMILIA should be understood as following: 2.2 Emilia believes that P and alieves that not-P. Gendler (2008) holds that beliefs are representational cognitive attitudes that are responsive to reasons and to evidence and, as such, follow rational rules. The dissonant a-behavior, however, does not seem to be related to any psychological state with these characteristics. She introduces the term ‘alief’ to refer to psychological states that, according to her, fall beyond our present psychological categories. In contrast to beliefs, aliefs are associative, automatic and arational patterns of responses (Gendler 2008: 641). According to her, an alief is a mental state with an associatively linked cluster of contents that tend to be co-activated (Gendler 2008: 642–643). In paradigmatic cases, an activated alief has three types of components: (a) a representational content that is not necessarily propositional or 9 conceptual; (b) the experience of some affective or emotional state; and (c) the readying of some motor routine.10 Features of the subject’s internal or ambient environment activate aliefs. They are neither reality-sensitive in the way that beliefs are nor reality-insensitive in the way that imagination is (Gendler 2008: 651). Applying Gendler’s approach to EMILIA, Emilia’s assertion expresses a belief while her a-behavior indicates an alief with approximately the following content: ‘Women. Not competent beings! Do not let them represent your interests.’ The cogency of Gendler’s proposal resides in her treatment of several examples that are somewhat similar to EMILIA. To give her account more plausibility, consider the following variations of the dissonance case: GRAND CANYON:11 Suppose that our character, Emilia, decides to take a trip to the Grand Canyon. She plans to visit the Grand Canyon Skywalk, a horseshoe-shaped cantilevered glass walkway that extends 70 feet from the canyon’s rim and sits 4,000 feet above the Colorado River. Midway through the ‘adventure’ of crossing the Skywalk, someone asks her: ‘Do you think this is really safe?’ She answers unhesitatingly, ‘Of course it is safe, there is no doubt about it’. However, Emilia cannot help but feel absolutely terrified while crossing the Skywalk. She holds onto the railing as tight as she can, she sweats more than usual, and when she reaches the end, she suddenly shouts out: ‘I did it!’ For Gendler (2008), it is clear that Emilia believes that the Skywalk is safe. She not only avows it, she also drove miles to get there, paid the ticket and so on. However in addition to that belief, there is something very different going on. She has an alief with the approximate content: ‘Really high up, long long way down. Not a safe place to be! Get off’ (Gendler 2008: 635). 10 11 Gendler (2008) emphasizes that this is only a tentative definition. This is a version of the opening example in Gendler’s seminal paper (2008) on aliefs. 10 WALLET12: Instead of going to the Grand Canyon, Emilia decides to take a plane to Las Vegas and meet her friends there. At the airport, Emilia realizes that she forgot her wallet at home, which terrifies her. In spite of that, she decides to continue with the trip. Arriving in Las Vegas, without any cash or credit cards, Emilia tells her friends the story and asks to borrow some money. With the cash in hand, she immediately searches for her wallet in her bag. She probably wanted to keep the money in a safe place, but she borrowed the money because she did not have the wallet with her! ‘How silly of me!’ Emilia thinks. It seems clear that Emilia has not completely forgotten that she had left her wallet at home. But it is also clear that such information was not part of her practical reasoning about where to put the borrowed money. Gendler would say that together with her belief that she left her wallet at home, there is a very different thing going on: an alief that is activated by her habits and routine with roughly the content, ‘Money. A valuable thing! Put it in the wallet.’13 As Gendler would like to stress, GRAND CANYON and WALLET are much less ‘politically disturbing’ than EMILIA, but no less perplexing. Moreover, they seem to be situations that all of us could find ourselves in. What is perplexing about them is how a conscious belief seems to have no power over what underlies one’s a-behavior. One possible explanation—not explicitly endorsed by Gendler—appeals to the idea of dual processing. In philosophy, but especially in the psychology literature, authors have postulated the existence of two kinds of systems for processing information. Roughly, the hypothesis holds there to be a distinction between ‘processes that are unconscious, rapid, automatic, and high capacity, and those that are conscious, slow and deliberative’ (J. Evans 2008: 256). Because 12 13 This is a version of Gendler’s (2008) example in which she herself forgets her wallet at home. I am ruling out the possibility that these cases involve a shift from believing that P to believing that not-P, for in all cases Emilia is able to endorse her judgment while behaving contrary to it. 11 they are different kinds of processes, following different patterns and functions, and somehow independent of each other, their outputs can sometimes be very divergent. Gendler’s approach claims such a divergence. Dual-processing theories have received support from some experiments describing situations very similar to dissonance cases. For example, in the experiment conducted by Epstein and Denes-Raj (1994) subjects made systematically non-optimal choices, even while knowing that the options they were choosing were the worst ones.14 There is a reasonable amount of data favoring the dual-processing hypothesis. But there is also a lot of disagreement about the cluster of attributes associated with both systems (J. Evans 2008). It is disputed whether being responsive to evidence is a characteristic of one system in particular. According to Irene Blair (2002), studies in recent years on the malleability of automatic stereotypes and prejudice have multiplied. These studies suggest that those automatic processes are not inflexible, inevitable, or uncontrollable. Such responses have been shown to be sensitive to a wide range of strategic, social, and contextual influences. For example, in the experiment conducted by Dasgupta and Greenwald (2001), people’s automatic racial attitudes were significantly affected by exposure to pictures of admired black people.15 14 The experiment presented subjects with an opportunity to gain money if they drew a red bean from a chosen bowl of mixed beans. In one of the cases, one bowl contained 1 red bean out of 10 beans and the other contained 9 red beans out of 100 beans. Even when provided with the percentages of red beans in each bowl, the great majority of participants chose to draw from the second bowl. Epstein and Denes-Raj (1994: 826) hold that such an experiment ‘contributes to accumulating evidence in support of the assumption of dual processing theories’. In contrast to what Nisbett and Wilson (1977) have been emphasizing, such experiments do not purport to show that our reasoning is full of irrational steps. They try to show that a rational response can be overwhelmed by the output of the other system, ‘even when people are fully aware that their resultant behavior is irrational’ (Epstein and Denes-Raj 1994: 826). 15 The experiment exposed subjects to pictures of admired black people and disliked white people (and viceversa). Their results revealed that such exposure ‘significantly weakened automatic pro-white attitudes for 24 hours beyond the treatment but did not affect explicit racial attitudes’ (Dasgupta and Greenwald 2001: 800). 12 If being responsive to evidence is one hallmark of belief, the notion of aliefs is not helpful in understanding what underlies one’s a-behavior. For our automatic and associative responses—as well as some of our habits—seem to change very quickly in response to new evidence. Sometimes, verbal avowals are less sensitive to evidence than automatic and associative responses (Schwitzgebel 2010: 539). As with Gendler’s notion of alief, a belief can also activate affective, emotional and a-behavioral reactions. Moreover, we could construct an analogue case of dissonance in which a person asserted a false irrational belief while her a-behavior was in fact very sensitive to the evidence.16 In this alternative case, what underlies her a-behavior is reality-sensitive, as Gendler’s concept of belief apparently requires.17 However, we are still left with an open question. In the cases at hand, Emilia’s abehavior is resistant to evidence, which seems to be a crucial point in generating Emilia’s dissonance. Therefore, it seems that besides identifying the psychological states that are in play in a dissonance case, we are forced to explore why such a-behavior is so inflexible, once we give up the idea that it is inflexible because it is guided by aliefs. I shall next explore this issue through my interpretation of EMILIA. 2.3 Emilia believes that P and believes that not-P. In the first subsection, I defended the view that even though Emilia’s assertion can be understood as a product of her judgment it should also count as the expression of a belief. Being committed to the truth of a proposition reached by reasoning from the evidence is 16 Imagine someone who endorses his community’s openly sexist values–such as women are less fit for politics than men–though at the same time he cannot avoid treating everyone in an egalitarian way. One possible explanation is that the mental state underlying the latter pattern of behavior responds correctly to the evidence that men and women are equally fit for politics. 17 Gendler (2008) would insist that a belief responds to reasons while what underlies a-behavior does not. Therefore, beliefs differ from aliefs. However, responsiveness to reasons seems to be one kind of responsiveness to evidence. 13 central to what it is to believe a proposition. This idea is compatible with the fact that not all beliefs are formed through reasoning and judgment. In the previous subsection, I argued against the approach that appeals to aliefs. We could have a dissonance case with the opposite pattern of EMILIA: one’s assertion being highly irrational and insensitive to the evidence while one’s a-behavior is very sensitive. These considerations suggest that in EMILIA, we are dealing with the presence of two contradictory beliefs. One of them is systematically resistant to the evidence and largely out of the subject’s control. However, we are now left with the question about the circumstances in which a single subject holds a pair of contradictory beliefs. Self-deception appears to involve holding contradictory beliefs. Although there is no consensus about its characterization, it seems that self-deception involves a degree of irrationality with respect to implementing evidence that the subject has. The controversy centers on explaining how the errant belief was formed. There are those who hold that the false belief is formed intentionally (see Davidson 1982, Pears 1984, Sartre 1956).18 This is the intuition behind the idea that the subject somehow deceives herself. There are others who hold that self-deception does not involve intentionality (see Johnston 1988, Mele 2001), apparently dissolving the paradox in which the subject is at the same time the liar and the victim. There are alternative explanations that argue that self-deception does not require the formation of contradictory beliefs (see Funkhouser 2009), or even extreme views which hold that self-deception does not involve beliefs but a different psychological attitude instead, such as pretenses (see Gendler 2007). In a sense, Emilia’s a-behavior could suggest that she is self-deceived: if there is a belief guiding it, it is a false belief that is resistant to her evidence. However, simply having a 18 I am referring to the so-called intentionalist theories of self-deception. 14 belief with such characteristics does not seem to be enough to classify Emilia as selfdeceived. After all, she keeps track of her a-behavior because of her conscious commitments. She is aware both that she is committed to the truth of P and that her abehavior conflicts with her commitment. In a sense, she tries hard not to deceive herself. As already noted by Gendler (2008) and Schwitzgebel (2010), self-deception misrepresents the structure of dissonance cases. Nevertheless, in having two contradictory beliefs, Emilia is still subject to some sort of irrationality. Emilia is irrational without being self-deceived. There is another relevant difference between a paradigmatic case of self-deception and a dissonance case: we are not interested in answering how anyone could form a belief that is contrary to her evidence (Pears 1984: 39), but instead, how anyone could maintain a belief that is contrary to her evidence. My suggestion is that the persistence of a false belief is due to a belief system malfunction. The false belief fails to update in response to new information partly because of the influence of deep linkages with other psychological states that help maintain an old and rejected belief state. In the normal case, if I change my mind about a given issue, such a change has consequences for other thoughts linked to it. If I come to know that the organoleptic characteristics of a wine have an intrinsic relation to the type and quality of the grape used to make it, I will stop expecting a 2005 bottle to be the same as a 2004 because the growing seasons were so different during those years. And normally, I do not need to be aware of all the inferential linkages that a new piece of information has to existing psychological states in order to change my other beliefs. My beliefs should adjust automatically to the new information or to the formation of a new belief. However, failures in such updating are also common. We can imagine a person learning the new information about wines, but 15 continuing to praise a certain wine for being exactly the same every year, as if both thoughts were not related to each other. Consider Paul, who has recently become a vegetarian. He cannot stand the taste of meat anymore. Since he was a child, his favorite dish has been goulash soup. Last week, Paul was invited to his friend’s house and he was asked to bring the tastiest dish that he could think of. The first dish that came to Paul’s mind was goulash soup. Goulash could not be the tastiest dish to him though, since he cannot stand the taste of meat. In this case, an old judgment about his preferred dish was not updated with Paul’s new beliefs and tastes. If we go back to EMILIA, we could think of it as an extreme case of failure in updating one’s beliefs. Emilia retains beliefs that have not been updated in response to the information acquired by reasoning from the evidence about each gender’s political capabilities. Even undertaking a drastic and conscious transformation of her view about the political abilities of men and women, Emilia cannot change her contradictory belief. It is likely that she cannot do so because the belief that women are less politically capable plays a central role in her psychology and is connected to so many other psychological states that changing it is very hard. We could imagine some different genealogies for such a belief. We could imagine that Emilia was raised in a community so deeply informed by sexist values that stereotypes about gender maintain Emilia’s old belief even when she tries to change it. In this situation, changing only one aspect of a whole worldview is not enough. She would need to drastically reform her beliefs. Alternatively, we could imagine that Emilia had such a traumatic experience trying to become a politician that maintaining the old stereotype that women are not made for politics provides comfort. When Emilia formed the belief that men and women are equal in politics, the past belief that they are not could not be eliminated because strong emotions held it in place. 16 To sum up, when Emilia reasons about the evidence regarding gender equality, she forms the belief that men and women are equal in their political competencies. However, this is not enough to effect a full change of mind. She ends up with mutually contradictory beliefs. In such a situation, instead of promoting a thoroughgoing change, Emilia acquires an extra belief. Her situation is such that other psychological states block her conscious attempt to fully change her mind and rid herself of her old contradictory belief. Cases such as EMILIA show that just discovering one’s irrationality does not suffice to fix it. A belief recognized as false may still play a central role in one’s psychology. In EMILIA, Emilia ends up with two contradictory beliefs because of some specific failure in updating her beliefs. 3. Answering Criticisms In this section I will consider two possible criticisms of my reading of the case. The first concerns its limits in explaining different dissonance cases. I argued that EMILIA is a case involving two contradictory beliefs. I attempted to make such an interpretation plausible by thinking of it as involving a belief that was not properly updated. However, perhaps someone would object that such a reading is not applicable to all dissonance cases. GRAND CANYON, for example, is a case in which it is not clear that Emilia’s a-behavior is linked to a false and irrational belief maintained by some failure in updating. I acknowledge such a concern. But it does not present a threat to my approach because I only need my reading to cover some of the relevant cases. A dissonance case is a case in which a person sincerely asserts that P, but a-behaves as if he or she believed that not-P. If my approach is right, a person in some dissonance cases abehaves as if he or she believed that not-P because he or she believes that not-P. This happens in 17 EMILIA. It also probably happens in WALLET, and in the case of Paul, who recently became a vegetarian. In those cases, it is plausible to say that some new information has not been properly used to update one’s belief system. However, this explanation allows that in some other dissonance cases, the subject a-behaves as if he or she believed that not-P because of the presence of a different psychological state. In fact, a pluralist view of dissonance cases, according to which different dissonance cases can express different psychological phenomena, seems to be the most accurate approach. Several psychological states such as an emotion or a perception could be underlying one’s a-behavior. Probably, in a limited number of cases, an a-behavior could even be a simple motor response without any psychological state underlying it. Psychological states other than a belief can guide one’s behavior. Concerning GRAND CANYON, maybe we need to refer to a different psychological state other than a belief to explain Emilia’s abehavior. The perception of the Grand Canyon through the glass is one option. Another option is Emilia’s visceral fear of heights.19 Such cases do not threaten my approach because my approach only requires that some relevant dissonance cases involve contradictory beliefs. EMILIA is such a case. A second possible criticism can be raised if we consider Schwitzgebel’s approach. According to him, if we consider Emilia as a whole, we should not ascribe both contradictory beliefs to her. Emilia should be described instead as being in an in-between state of belief regarding the political competences of men and women. Schwitzgebel holds a dispositionalist account of belief; an account that ‘equates believing with being disposed to act and react in various ways in various circumstances’ 19 Considering such options does not lead me back to Gendler’s approach. Once we use finer distinctions among different psychological states, we do not need to appeal to a new notion, such as aliefs, to play the role of an umbrella concept to cover different cases of a-behavior. 18 (Schwitzgebel 2010: 533). According to him, to believe is to possess ‘a cluster of dispositions, which can include cognitive and phenomenal dispositions as well as behavioral ones’ (Schwitzgebel 2010: 535). In EMILIA, Emilia possesses only some of the relevant dispositions to count as believing that men and women are equally competent in politics. She has also some of the relevant dispositions to count as believing its negation. In certain contexts, the attribution of either belief to her would seem accurate. For example, a colleague in her discussion group on sexism might say that Emilia is not sexist. However, according to Schwitzgebel, if we consider her dispositions as a whole, it is not quite accurate to ascribe either belief to Emilia, nor is it accurate to say that Emilia lacks each belief. It is questionable whether a dispositionalist account can satisfactorily explain what a belief is. However, even if we do not enter that dispute, Schwitzgebel’s account gives rise to problematic consequences concerning Emilia’s understanding of her dissonance. In describing what her dissonance consists in, Emilia would be forced to answer that she ‘kind of’ (Schwitzgebel 2010: 547) believes in the equality of the genders. Such an account fails to explain the obvious irrationality and conflict in her psychological state. A dissonance case such as EMILIA indicates that there is something deeply wrong with the subject. It is not only a matter of ‘kind of’ believing something. Possibly, Schwitzgebel’s notion of an in-between state of belief represents some type of psychological conflict. However, it still misrepresents the case. Once the dissonant subject learns about herself in terms of in-between believing, she is forced to change her conscious commitments. However, the problem in a dissonance case is not that the subject has a poor understanding of her conscious commitments, but that such commitments have no control over what underlies her a-behavior. Emilia is consciously committed to the truth of the equality of the genders, but some other states maintain her false belief that they are unequal. 19 That false and irrational belief is then resistant to Emilia’s conscious commitments. Emilia’s problem is not that she does not believe in the equality of the genders, but that there is something dissonant in her attitudes. Attributing contradictory beliefs to Emilia does not ignore her overall situation. The fact that Emilia is irrational is her real problem.20 In sum, the contradictory-belief view of cases like EMILIA defended in the last two sections contrasts with the approaches offered by Gertler, Gendler and Schwitzgebel. Gertler would deny the presence of one of Emilia’s beliefs––the one linked with her explicit endorsements. Gendler would deny the presence of the other belief––the one linked with her a-behavior. And finally, Schwitzgebel would deny the presence of both beliefs in Emilia’s psychology. 4. Dissonance and Moore’s Paradox Section 1 presented an argument according to which, given a certain interpretation of Emilia’s dissonance, she is in a position to assert (and believe) a Moorean proposition once she learns about her dissonance. In the last sections, I defended such an interpretation. I argued for the view that Emilia has contradictory beliefs. In doing so, I presented a case in which Emilia is in a position to assent to P at the same time that she acknowledges that she believes that not-P. However, given that Moorean propositions are considered absurd things to assert or believe, the proposed reading of the case appears to lead us to a puzzle. It seems that Emilia could not reasonably assert or believe a Moorean proposition. In this section, I briefly discuss some views on Moore’s paradox that attempt to explain why a Moorean proposition is absurd. My aim in this section is to present some reasons to dissipate the assumption that 20 See Borgoni (2014a) for an extended analysis of Schwitzgebel’s account. 20 any Moorean proposition, like the one Emilia is in the position to hold, is absurd and thus, unreasonable and unjustified. A Moorean proposition is a proposition that can be expressed by a Moorean sentence. Moorean sentences come in two forms: (i) ‘P but I do not believe that P’ and (ii) ‘P but I believe that not-P’.21 Given this paper’s working case, I focus on (ii). It is a shared assumption that asserting or believing a Moorean proposition is absurd, although it might accurately describe a state of affairs. The sun can be shining in Los Angeles at the same time that a given person, let us say Louis, believes it is not. But Louis’ assertion (or belief) ‘The sun is shining in Los Angeles but I believe it is not’ is supposedly absurd. It does not involve any kind of logical error as would be involved in: ‘The sun is shining and the sun is not shining’. Still, it seems to be an absurd thing to say (Williams 2006) or to believe (Shoemaker 1996). The solution to Moore’s paradox lies in the explanation of such absurdity. There are various competing approaches to this matter.22 One line of explanation holds that a Moorean proposition involves some implicit contradiction. For example, one view argues that whenever one asserts that P, one implies23 that he or she believes that P (see Moore 1942, 1944). This, however, does not yield a contradiction for (ii). In affirming ‘P but I believe that not-P’, Emilia manifests her contradictory beliefs, but not a belief in a contradiction. Having contradictory beliefs might be an irrational state of mind (Davidson 1982), but not an absurd one. This line of explanation is compatible with Emilia being in a position to assert or believe a reasonable Moorean proposition. 21 Cf. Williams (1979). Moore’s (1942) original sentences have the form ‘I believe that P but not-P’ (‘I believe he has gone out, but he has not’), and ‘P but I do not believe that P’ (‘I went to the pictures last Tuesday but I don’t believe that I did’). 22 See Green and Williams (2007) for an overview of available approaches. They sketch eighteen different attempts to explain Moorean absurdity. Such approaches refer to a severe failure either of theoretical rationality or of practical rationality. 23 ‘…in an everyday sense of “imply”’ (Moore 1942: 542). 21 A different view argues that an implicit contradiction between P and not-P derives from the purported fact that saying that one believes that P is just another way of saying ‘P’ (see Wittgenstein 1953, Linville and Ring 1991, Heal 1994).24 If this principle were correct of all belief self-ascriptions, then Emilia would be in a position to assert (and believe) a contradiction. However, ‘I believe that P’ and ‘P’ do not always have the same use. Very often, one says ‘I believe that P’ instead of plainly saying ‘P’, to express a lower degree of confidence in the truth of P. More important to the current discussion, one might also use a belief self-ascription to tell the hearer something about his or her mind that contrasts with what one consciously holds.25 Emilia could self-ascribe the belief that not-P to tell her therapist, for example, that she has a recalcitrant belief. Because the ascribed belief is a recalcitrant one, of which she disapproves, such a belief self-ascription ‘I believe that not-P’ does not have the same use as ‘not-P’. A third line says that the problem lies at the level of beliefs and not at the level of assertions (see Shoemaker 1996, Kriegel 2004): to believe a Moorean proposition is impossible, or ‘at least, logically defective’ (Shoemaker 1996: 76). Therefore, the subject cannot coherently assert such a proposition. According to this line of reasoning, believing a Moorean proposition is logically defective because in believing it, the subject believes a contradiction. However, again, for propositions of form (ii), it is not clear that the subject believes a contradiction. 26 In Emilia’s case, for example, her recalcitrant belief in not-P is such that she does not assent to the content of the belief. Although she might be in a 24 This approach seems to be insufficient to deal with (i). ‘I do not believe that P’ is not another way of saying ‘not-P’. 25 See Hunter (2011) for cases of alienated beliefs. 26 In the revised version of his paper, Shoemaker holds that it is impossible to believe the content of (i), but not the content of (ii). 22 position to assent to ‘I believe that P and I believe that not-P’, she is not in a position to assent to not-P, and thus is not in a position to assent to ‘P and not-P’. Another popular view is that a Moorean assertion or belief is absurd because it represents a deep failure of self-knowledge (see Moran 2001, Shoemaker 1996). From the first-person perspective, assenting to I believe that P has an intrinsic connection with assenting to P. Both if I sincerely assert that P or that I believe that P, I reveal my conscious commitment to the truth of P. Evans’ (1982) remarks on the special way we self-ascribe mental states indicate that there must be some principle of coordination between the two attitudes: ‘whenever you are in a position to assert that P, you are ipso facto in a position to assert “I believe that ‘P”’ (Evans 1982: 225-226).27 If the coordination fails, one is likely irrational. The present paper does not pursue an account of Moore’s paradox. However, it is compatible with the view that the absurdity of a Moorean proposition has something to do with the way we ascribe beliefs to ourselves from the first-person perspective. Usually, if I discover that the sun is shining, I just change my mind and give up my previous belief that the sun is not shining. Normally, my self-ascription of such a belief automatically accompanies the change in my first-order belief. From the first-person perspective, I do not need to examine any evidence about my psychological states to self-attribute them. However, not all beliefs are accessible from the first-person perspective. Unconscious beliefs are an example of such a case. Moreover, not all beliefs change once the subject gains evidence against them or new beliefs, like in Emilia’s case.28 27 I call this phenomenon a ‘principle of coordination’ instead of ‘transparency’ to maintain a weaker interpretation of Evans’ remarks. An entire paper would be necessary to deal with the increasing number of criticisms of transparency. 28 Recognizing such a feature of our psychology defies another attempt to explain the absurdity of Moorean propositions in epistemological terms. Klein (1986) and De Almeida (2001) hold a rule of revision of beliefs 23 As suggested previously, Emilia grasps her own dissonance from the third-person perspective on herself. She will ascribe mental states to herself by reasoning from the evidence about them. This set of evidence is different from the set of evidence she relies on to assert that men and women are equally competent in politics. Potentially, she can arrive at two contrasting judgments as to whether P and whether she believes that P. It is because Emilia has a recalcitrant belief that she cannot change her belief that not-P even though she regards P as true. A case such as EMILIA is perplexing because the subject is in a state of psychological conflict. However, it seems reasonable to accept that Emilia’s psychological conflict—a state of irrationality as I suggested—could be available to her. If it turns out that the correct description of her irrationality makes a Moorean proposition available to Emilia, the very assertion of the proposition should not be irrational. A Moorean sentence is a strange one, but a subject could be justified in asserting one despite its strangeness. For part of the Moorean proposition is available to Emilia by reasoning the way others do when coming to know about another’s mind. In fact, being strange or absurd does not seem to be a proper criterion to identify an assertion as unjustifiable. Given Emilia’s irrationality and her view on the subject matter, asserting or believing a Moorean proposition is reasonable. In this sense, Emilia is not prevented from being justified in asserting a Moorean proposition due to the risk of being irrational in doing so. 5. Grasping the Dissonance and Asserting a Moorean Proposition according to which ‘when a proposition P is added to a belief system, any belief that would block the warrant path to that belief must be removed from the belief system’ (cf. Green and Williams 2007). According to the rule, Emilia’s belief that P is a reason to refrain from believing not-P, which seems correct. However, as I have been arguing, some dissonance cases like EMILIA are particularly interesting because the conflicting belief is resistant to the reasons against it. The rule of revision is at most a normative, not descriptive, claim about our belief system. 24 I have argued that some relevant dissonance cases are cases in which the subject has two mutually contradictory beliefs. This description is slightly different from the one hypothesized in Section 1. There, the dissonance was characterized as a conflict between a mere assertion and a belief. However, a fuller diagnosis is given in terms of contradictory beliefs.29 Let ‘P’ be the content of Emilia’s assertion, men and women are equally competent in politics. Emilia, who sincerely and with conviction asserts that P but a-behaves as if she believes that not-P, has two mutually contradictory beliefs. She consciously believes that P, but, as manifested by her a-behavior, also believes that not-P. Given the characteristics of the third-person perspective on oneself, that description of her dissonance is available to Emilia. She is able to discover what underlies her dissonance by reasoning from the evidence about her mental states in the same way we did. Consequently, she is in a position to acknowledge ‘I have two mutually contradictory beliefs: I consciously believe that P but, as my a-behavior indicates, I also believe that not-P’. The root of Emilia’s dissonance lies in her incapacity to get rid of her deeply entrenched belief that not-P. This belief contrasts with her conscious and sincere commitments. Emilia’s beliefs are such that although she has both, and is justified in selfattributing them, she asserts that P while she does not assert that not-P. Learning that she has the belief that not-P does not support her asserting that not-P. Therefore, given Emilia’s conscious commitment to P and what she has learned about her dissonance, Emilia is in a position to assert: ‘P, and I believe it, but I also believe not-P’. If Emilia can assert this 29 My view differs from Sommers (2009). He holds that dissonance cases involve dissonant but not mutually contradictory beliefs. He argues for the distinction between ‘de mundo’ and ‘de dicto’ beliefs. My view also differs from Bilgrami (2006). Bilgrami would probably agree that Emilia has two beliefs, but his account depends on the distinction between two kinds of beliefs: purely dispositional beliefs and beliefs as commitments. In contrast with both accounts, according to my view, Emilia’s beliefs are of the same type of psychological state whose contents have the same structure. 25 variation of a Moorean sentence, she can also assert a Moorean sentence in its original form, ‘P, but I believe not-P’, by removing the middle conjunct.30 Emilia is justified in asserting this type of Moorean proposition because of her reasons for assenting to P and because of what she has learned about herself. She has reasonable evidence both in favor of P and in favor of the fact that she has two mutually contradictory beliefs. A dissonance case, such as EMILIA, is a special situation in which a Moorean proposition is both natural and justified once the dissonant person reasons about the evidence on the subject matter and on herself. The result that a dissonant person is in a position to assert a Moorean proposition is not exclusive to this paper’s reading of dissonance cases. This paper’s position is that it is a sufficient condition for allowing special occurrences of justified Moorean assertions (and beliefs) that the contradictory-belief view is an accurate reading of dissonance cases. The reading based on Gertler’s (2011) view, for example, also predicts that the dissonant person can reasonably assert a Moorean proposition.31 According to such a view, since the person judges that P while acknowledging to have a dispositional belief that not-P, he or she is in the position to hold ‘P but I believe that not-P’ (See Gertler 2011). However, in contrast with the above competing approaches, this paper’s view preserves the connection between Moorean propositions and the existence of some related 30 I thank an anonymous reviewer for elaborating this step. It is an open question whether a particular class of dissonance cases could put the subject in the position of asserting, and believing, a legitimate omissive form of Moorean propositions like ‘P, but I do not believe that P’. If there is a case in which the individual judges a proposition while noticing that he or she does not believe it, that would be such a case. I also thank an anonymous reviewer for the articulation of this possibility. However, this class of dissonance cases––if they exist––differs from EMILIA in important ways, and it is not within the scope of this paper. Cases like EMILIA are paradigmatic examples of dissonance cases. 31 Gendler and Schwitzgebel’s views do not have any such consequences. According to Gendler, the dissonant person believes that P while alieving that P. Thus, when the person combines his view on P and on his psychology, he would say ‘P, and I believe so, but I also alieve that not-P’. According to Schwitzgebel’s approach, the person would be in a position to affirm ‘P, but I kind of believe that P’. In neither case does the person assert a Moorean proposition. 26 phenomenon of irrationality. For the contradictory-belief view acknowledges that dissonance cases like EMILIA involve a type of irrationality that underlies the subject’s dissonance. The dissonant subject maintains two mutually contradictory beliefs. However, the challenge of this paper’s position to the orthodox view on Moorean propositions is that, in grasping his or her own dissonance, the subject is not irrational.32 Once the dissonant person combines her conscious commitments with the description of her dissonance in terms of mutually contradictory beliefs, the person is in a position to affirm (and believe) a Moorean proposition. Such a proposition is justified by the evidence the person has in support of P and by the evidence she has in support of her having two mutually contradictory beliefs.33 Cristina Borgoni University of Graz REFERENCES Blair, I. V. 2002, ‘The Malleability of Automatic Stereotypes and Prejudice”, Personality and Social Psychology Review 6 (3), pp. 242–261. Bilgrami, A. 2006, Self-Knowledge and Resentment, Cambridge, Mass: Harvard University Press. Borgoni, C. 2014a, “Dissonance and Irrationality: A criticism of the in-between account of dissonance cases”, Pacific Philosophical Quarterly DOI: 10.1111/papq.12039. Online first. 32 What I call ‘the orthodox view’ holds that Moorean beliefs and assertions are themselves irrational. This view can be found in Shoemaker (1996). However, there are signs that the orthodox view is evolving to become more consistent with the view defended in this paper, namely, that Moorean beliefs and assertions indicate irrationality in the believer or asserter. Moran (2001) seems to hold the latter thesis. 33 Thanks to Justin Bledin, David Bordeaux, Andrew Hsu, Peter Murray, Peter Myrdal, Calvin Normore, Manuel de Pinedo, and Hong Yu Wong for valuable conversations and helpful comments on previous drafts. I’m also grateful to Lucy O’Brien and Tyler Burge for numerous discussions and suggestions on the various stages of this work. This research was partly funded by the research projects ‘Information and the Fragmented Mind’ (granted by the State of Styria), ‘The Fragmented Mind’ (granted by the FWF Austrian Science Fund), and FFI2013-44836-P (granted by the Spanish Ministry of Economy). 27 Borgoni, C. 2014b, “Dissonance and Doxastic Resistance”, Erkenntnis DOI 10.1007/s10670014-9691-0. Online first. Cassam, Q. 2010, “Judging, Believing and Thinking”, Philosophical Issues 20 (1), pp. 80–95. Chislenko, E. 2014, “Moore’s Paradox and Akratic Belief”, Philosophy and Phenomenological Research DOI: 10.1111/phpr.12127. Online first. Dasgupta, N. and Greenwald, A. G. 2001, “On the Malleability of Automatic Attitudes: Combating Automatic Prejudice with Images of Admired and Disliked Individuals”, Journal of Personality and Social Psychology 81 (5), pp. 800-814. Davidson, D. 1982, “Paradoxes of Irrationality”, in: 2004, Problems of Rationality, Oxford: Clarendon Press, pp. 169-88. Davidson, D. 1984, “First Person Authority”, in: 2001, Subjective, Intersubjective, Objective, Oxford: Clarendon Press, pp. 3-14. De Almeida, C. 2001, “What Moore’s Paradox is About”, Philosophy and Phenomenological Research 62: 33–58. Epstein, S. and Denes-Raj, V. 1994, “Conflict Between Intuitive and Rational Processing: When People Behave Against Their Better Judgment”, Journal of Personality and Social Psychology 66 (5), pp. 819-829. Evans, G. 1982, The Varieties of Reference, New York: Oxford University Press. Evans, J. 2008, “Dual-Processing Accounts of Reasoning, Judgment, and Social Cognition”, Annu. Rev. Psychol. 59, pp. 255-278. Funkhouser, E. 2009, “Self-Deception and the Limits of Folk Psychology”, Social Theory and Practice 35, pp. 1-13. Gendler, T. 2008, “Alief and Belief”, Journal of Philosophy 105, pp. 634-663. Gendler, T. 2007, “Self-deception as Pretense”, Philosophical Perspectives 21, pp. 231-258. 28 Gertler, B. 2011, “Self-Knowledge and the Transparency of Belief”, in: A. Hatzimoysis, ed., Self-Knowledge, Oxford: Oxford University Press, pp. 125-145. Green, M. and Williams, J. N. 2007, “Introduction”, in: Green, M. and Williams, J. N., eds., Moore’s Paradox: New essays on Belief, Rationality, and the First Person, Oxford: Oxford University Press, pp. 3-36. Heal, J. 1994, “Moore’s Paradox: A Wittgensteinian Approach”, Mind 103, pp. 5-24. Hunter, D. 2011, “Alienated Belief”, Dialectica 65 (2), pp. 221–240. Johnston, M. 1988, “Self-Deception and the Nature of Mind”, in: McLaughlin, B. P. & Rorty, A. O., ed., Perspectives on Self-Deception, Berkeley and Los Angeles: University of California Press, pp. 63-91. Klein, P. 1986, “Immune Belief Systems”, Philosophical Topics 14: 259–80. Kriegel, U. 2004, “Moore’s Paradox and the Structure of Conscious Belief”, Erkenntnis 61, pp. 99-121. Linville, K., and Ring, M. 1991, “Moore’s Paradox Revisited”, Synthese 87: 295–309. Mele, A. 2001, Self-Deception Unmasked, Princeton: Princeton University Press. Moore, G. E. 1942, “A Reply to My Critics”, in: P. Schilpp, ed., 1942: The Philosophy of G. E. Moore, La Salle: Open Court, pp. 535-677. Moore, G. E. 1944, “Russell’s ‘Theory of Descriptions’”, in: P. Schilpp, ed., The Philosophy of Bertrand Russell, Evanston: Northwestern University, pp. 175-226. Moran, R. 2001, Authority and Estrangement: An Essay on Self-Knowledge, Princeton: Princeton University Press. Nisbett, R. E. & Wilson, T. D. 1977, “Telling More Than We Can Know: verbal reports on mental processes”, Psychological Review 8, pp. 231–259. 29 Peacocke, C. 1998, “Conscious Attitudes, Attention, and Self-Knowledge”, in: C. Wright, B. Smith and C. MacDonald, ed., Knowing Our Own Minds, Oxford: Clarendon Press. Pears, D. 1984, Motivated Irrationality, Oxford: Clarendon Press. Pruss, A. 2011, “Sincerely Asserting What You Do Not Believe”, Australasian Journal of Philosophy 90 (3), pp. 541-546. Sartre, J. P. 1956, Being and Nothingness (trans. Hazel E. Barnes), New York : Citadel Press. Schwitzgebel, E. 2010, “Acting Contrary to Our Professed Beliefs, or The Gulf Between Occurrent Judgment and Dispositional Belief”, Pacific Philosophical Quarterly 91 (4), pp. 531-553. Schwitzgebel, E. 2011, “Self-Ignorance”, in: JeeLoo Liu and John Perry, eds., Consciousness and the Self, Cambridge: Cambridge Press. Shoemaker, S. 1968, “Self-Reference and Self-Awareness”, in: Q. Cassam, ed., 1994, Selfknowledge, Oxford: Oxford University Press, pp. 80-93. Shoemaker, S. 1996, “Moore’s Paradox and Self-Knowledge”, in: The First-Person Perspective and Other Essays. Cambridge University Press, pp. 74-93. Sommers, F. 2009, “Dissonant Beliefs”, Analysis 69 (2), pp. 267-274. Turri, J. 2010, “Refutation by Elimination”, Analysis 70 (1), pp. 35-39. Williams, J. N. 1979, “Moore’s Paradox: One or two?”, Analysis 39, pp. 141–142. Williams, J. N. 2006, “Wittgenstein, Moorean Absurdity and its Disappearance from Speech”, Synthese 149, pp. 225-54. Wittgenstein, L. 1953, Philosophical Investigations (trans. G.E.M. Anscombe), New York: Macmillan. Wright, C. 1998, “Self-Knowledge: the Wittgensteinian Legacy”, in: C. Wright, B. Smith and C. MacDonald (ed.) 1998, Knowing Our Own Minds, Oxford: Clarendon Press, pp. 13-45. 30