Knowledge is not always the same as knowledge: there are different sorts of knowledge, which are far from equivalent psychologically […] The doctor’s knowledge is not the same as the patient’s and cannot produce the same effects. If the doctor transfers his knowledge to the patient as a piece of information, it has no result […] The patient knows after this what he did not know before—the sense of his symptom; yet he knows it just as little as he did.
Sigmund Freud, 1966, p. 348
The quest for self-knowledge has captivated philosophers, psychologists, and everyday folk for centuries. It is a journey that goes beyond simply knowing your favorite color or food, for example. It is about understanding the essence of who you truly are, your inner core, or even your very existence. As gripping as that might sound, it is not what this book is about (at least not primarily). I am not concerned with such a hard-to-get kind of self-knowledge or with self-discovery or self-exploration per se. Following recent debates in philosophy, I am, instead, mainly concerned with the kind of self-knowledge that is thought to be privileged, authoritative, particularly reliable, or even infallible. Interestingly, this self-knowledge is also often so easy to come by that it appears trivial to some.1 In gist, it is about knowledge of your occurrent beliefs, desires, emotions, and sensations. It is not about a kind of knowledge of your Self—if the Self is considered something that goes above and beyond your mental states.
There is, however, no need to sell this seemingly mundane kind of self-knowledge short. Indeed, what makes it seem trivial to some concurrently makes it fundamental. In an important way, the kind of ‘mundane’ knowledge I have of my belief that I am wearing socks is the ground for successful self-discovery, enlightening navigation through life, making life-altering decisions, and a deep understanding of my mental states. To put it in broadly Freudian terms, a doctor’s knowledge of a patient’s mind is different from the patient’s own knowledge, and cannot produce the same effects. The patient can come to know everything the doctor knows about her mind, and yet something important could be missing. That missing part can have different effects, and it is what the privileged kind of self-knowledge is about. This is the kind of knowledge I have of my beliefs in believing this or that, my emotions in feeling this or that way, or my desires in wanting something to become the case. Certainly, a main goal of therapy is self-knowledge. And a significant part of the therapist’s work lies in understanding, in coming to know the patient’s mind, by means of observation, inference, and interpretation. But for this kind of knowledge, and for therapy in general, to have the intended result, the patient must (also) be in the position to first-personally know her own mental states; to know her mental states in a way that is not available to others. What makes this sort of knowledge distinctive (more on this later) cannot just be passed on from the therapist to the patient. It would be one thing for Freud’s “Rat Man”2 to find out in therapy that he behaves in certain ways, has obsessive fears of rats, and suffers from anxiety owing to a repressed death wish toward his father. It would, however, be another thing for him to wish for his father’s death and know of that wish authoritatively and immediately without the need for observation, interpretation, or inference. Viewed this way, the privileged, first-personal, authoritative kind of self-knowledge missing in the Rat Man case (or more generally in everyday experiences of feeling disconnected from one’s own beliefs, emotions, or desires, involving a lack of ownership that usually goes along with being in one’s own mental states) appears far from trivial. First-personal self-knowledge is, in fact, crucially important for our mental health, and constitutive of a non-alienated self-relation to one’s agency and freedom.
Relatedly, a supposedly trivial kind of self-knowledge of my present mental states becomes strikingly significant when focusing on the occurrent beliefs, emotions, or deep desires that are usually considered to make up who I (truly) am. These include ethical, political, or religious beliefs (and not my belief that I am wearing socks). For a skeptic, the content of the specific belief, rather than the way in which one knows the belief, seems trivial in this example of knowing that I am wearing socks. However, to be fair, the skeptic’s point is stronger and more extensive: he objects that the realm in which our self-knowledge is privileged is limited to occurrent mental states, such as the belief that I am wearing socks (or perhaps that it is raining outside). It does not generalize to standing or dispositional beliefs that encompass those beliefs and are considered more significant and self-defining (e.g., one’s belief in racial equality or that God exists). As will become clear in later chapters, my disagreement with objections about limited applicability and lack of substance relies heavily on a different understanding of our mental states (according to which there exists no substantial difference between occurrent and dispositional mental states). In fact, the same can be said for the entire approach to self-knowledge I am going to defend. Thus, just as much as this book is about self-knowledge, it is also about the nature of our minds.3
Note also that this seemingly trivial self-knowledge deserves our philosophical attention precisely because it is not considered a mere private undertaking in the hope of self-healing or self-improvement. This kind of (trivial) self-knowledge has proven to be a complex and contentious issue. Indeed, various philosophical traditions and theories have offered distinct perspectives on what it means to have privileged self-knowledge. Nevertheless, although I disagree with Cassam’s objection that mainstream philosophical accounts of self-knowledge focus on trivial self-knowledge and with his inferentialism about self-knowledge (more on this later), I do not intend to dismiss or minimize the significance of acquiring self-knowledge through self-observation, interpretation, or testimonial evidence. It is not only in rare cases that self-knowledge takes the form of a self-attribution of mental states from a third-person stance, on grounds of self-observation (e.g., when you are uncertain about whether you really feel a certain way or when you realize that your behavior does not align with your beliefs); nonetheless, as we shall see, spectatorial self-knowledge does not make self-knowledge peculiar when compared with other forms of knowledge. And it is this difference, I believe, that sparks our “human” interest.
Scholars often take for granted that we can know our own minds in a way that is different from the way we, for instance, know about train cancellations, tables, and the minds of other persons. Nevertheless, the character, depth, and scope of this distinctiveness are disputed issues. One intuitively appealing and historically prominent answer is this: “I know what I believe, want, or feel by turning to the inside—away from the world around me.” In this book, I pursue the opposing line of thought: “If I want to know my own mind, then I should look outward onto the world rather than inward.” This approach follows a trend in the literature on self-knowledge, one that is usually attributed to Gareth Evans who famously claimed that “in making a self-ascription of belief, one’s eyes are, so to speak, or occasionally literally, directed outward-upon the world” (1982, p. 225). Evans argued that we can self-ascribe beliefs by directly considering their content. His observation has led not only to various versions of the so-called transparency view of self-knowledge but also to several pressing questions. One such question is a central motivation behind this book: the scope question.
Even if we agree about the transparency of beliefs, how can we possibly apply this to mental states other than belief? How could I possibly know my phenomenal states by only attending to the world around me? It seems very plausible that we can answer the mind-directed question (“Do I believe that p?”) by answering a corresponding world-directed question (“Is it the case that p?”). However, it seems significantly less plausible (or even counterintuitive) when it comes to our desires and emotions, let alone our sensations. We do seem to have some sort of first-personal awareness of these other mental states. Broadly speaking, there are two ways to go from here: (a) we can claim that this awareness is different from the kind of awareness I have of my beliefs or (b) we try to show that the transparency view of self-knowledge can be extended to mental states other than belief. I shall undertake the latter. I argue that the so-called transparency view of self-knowledge can and should be generalized beyond beliefs (or belief-like states) to our phenomenal states. But, before doing so, I should say a bit more about the explanandum and the arc of Feeling Outwards.
1 Knowing Me, Knowing You
Philosophical inquiry into self-knowledge’s distinctiveness usually begins by contrasting self-knowledge with other knowledge forms. This is not only meant to emphasize the relevance of self-knowledge but is also a common way to identify self-knowledge’s explanandum. More precisely, scholars treat the question of self-knowledge (understood in this standard way) as tantamount to the question of self–other asymmetry.4 There are different ways of describing and understanding this question, but it is generally framed as a question about differences between knowledge of my own mental states and the knowledge I have of others’ mental states.5
In this spirit, if I tell you that I believe that the restaurant next to my house has the best falafel, you will normally take what I say about my belief to be true. Normally, you will not check whether there is evidence contradicting, calling into question, or supporting my self-ascription. You are likely to simply take for granted what I say. And, if you and I disagree about whether I (really) believe this or desire that then I am likely to have the last word. This is most conspicuously the case if we disagree about what I feel. But when I ascribe a mental state to another person, I must do so on grounds of evidence gained through observation, inference, or testimony. I usually do not need to do this when ascribing these states to myself. It would be inappropriate to request that I justify my self-ascriptions or clarify how I have come to know the mental state I self-ascribe. When I sincerely report to someone that I am in pain, then that person generally takes my word on the matter because there is (usually) no need for me to ground my self-ascription (“I am in pain”) in any kind of evidence. In the most general terms, our present-tense self-ascriptions (avowals) differ considerably from our ascriptions of mental states to others. This is because the former are usually immediate (i.e., not based on inferential, perceptual, or behavioral evidence).
Assuming they are sincere, immediate self-ascriptions are also less prone to errors and thereby more likely to be correct. Consequently, we seem to enjoy some sort of authority when it comes to self-ascriptions—an authority that we lack in our mental state ascriptions to others. However, the fact that avowals are considered immediate and authoritative need not imply that we are infallible or omniscient about our own mental states. Obviously, we do not immediately know all of our mental states. In psychoanalysis, a whole therapeutic school and theory of the mind is based on the assumption that we have unconscious beliefs, feelings, and memories that we are typically unaware of and that we can learn or find out about in therapy, for example. According to the often-used iceberg metaphor for Freud’s theory of the unconscious, only the tip of the iceberg represents conscious awareness. A much larger portion—the unconscious mind—is submerged beneath the water. Regardless of how extensive we believe the unconscious to be (and regardless of the plausibility of Freud’s approach), no one will deny its existence.
Literature and movies are filled with cases of unconscious mental states that come to life through the characters’ seemingly inexplicable behaviors or motivations. These are often portrayed or interpreted as originating from repressed memories, desires, or anxieties. Besides our lack of privileged self-knowledge in cases of unconscious mental states, we also sometimes lack privileged knowledge of the mental states we are typically aware of. An example are cases of implicit (often racist or sexist) biased beliefs that influence our actions even while we sincerely assert the opposite. However, knowing our mental states also in the “third-personal” way—in a way that is also open to others: based on observation, interpretation, and inference—does not necessarily result in rejecting the previously mentioned asymmetries. It also does not necessarily entail adopting a Rylean view that reduces the asymmetry to the fact that I spend more time around myself and therefore possess (at least sometimes) superior behavioral evidence.6 As will become clear, this would not do justice to the fact that we typically know most of our mental states (beliefs, desires, intentions, emotions, and sensations) in an immediate way without drawing on evidence supporting that knowledge.
If we agree that self-knowledge deserves our attention, then we can take different routes to answering the question of what renders self-knowledge interesting enough to look into it. I am going to defend a view that departs from both Cartesianism on the one end of the spectrum and behaviorism à la Ryle on the opposite end. In doing so, I shall assume an asymmetry between self-knowledge and other-knowledge, one that does not presuppose a mysterious inner eye (or internal self-scanning process).7
Most contemporary philosophers reject a Cartesian view of introspection, which presents our mental states in such a way that we know of them without any doubt. Nonetheless, many of these philosophers try to hold onto the idea that there is something special about self-knowledge, and they do so in several different ways. Different accounts offer different explanations for the aforementioned asymmetry, and some offer more substantial explanations than others. The immediacy claim, namely the claim that a self-ascription is non-inferential, not based on evidence or that it is groundless, or transparent, for example, either just points to the effortlessness of avowals or amounts to a claim about the status of self-knowledge and how that kind of knowledge is acquired. The latter implies (a) that we can come to know our own mental states without inference (via some distinctive method of knowing unique to self-knowledge) or (b) that we directly know our mental states by being in the mental states in question.
To locate this book on the map of current approaches, it helps to follow a distinction drawn in the current literature—a distinction between epistemic and non-epistemic accounts of self-knowledge (see Gertler, 2008; Bar-On, 2004). The “epistemic” category here encompasses two types of accounts:
1. Accounts that take self-knowledge to be substantially different from other-knowledge in that it is achieved through some unique epistemic method.
2. Accounts that take self-knowledge to be gained by the same means as other-knowledge but consider the resulting self-knowledge to be more reliable.
The latter type are still epistemic because they give an answer to the question of how we find out about or detect our existing mental states and thereby acquire particularly reliable or secure knowledge.8 Proponents of the epistemic view maintain that our avowals’ first-person authority stems from our epistemically privileged position.
The term “non-epistemic” is somewhat misleading because, in a basic sense, any explanation of self-knowledge must be “epistemic.” However, most accounts that are said to belong in this category are epistemic in the minimal sense that self-knowledge is their explanandum. According to scholars in this camp, the difference between avowals and other mental state ascriptions is not a difference between epistemic mechanisms of discovery. Instead, in non-epistemic accounts, there is an assumption that self-knowledge and other-knowledge differ, but that this difference is not explained epistemically. The peculiarity of avowals—their authority and immediacy—is not grounded in some epistemically privileged access the subject has to her own mental states. It is also not grounded in her access to more, or a different kind of, evidence. It is, instead, explained through either the way in which the avowal is uttered or the nature of the mental state that is being avowed. Contrary to epistemic accounts, non-epistemic accounts either point to the expressive dimension of avowals (Bar-On, 2004; Finkelstein, 2003) or to the metaphysics of our mental states (Shoemaker, 1996; Moran, 2001, 2012; Bilgrami, 2006; Boyle, 2009, 2011, 2024).9
The views categorized as non-epistemic share the assumption that an adequate account of self-knowledge must take into account a non-epistemic dimension. However, they seem to vary when it comes to the extent of that assumption. Some of these views are epistemic in the minimal sense that their explanandum is commonly called self-knowledge and thus considered an epistemological topic.10 Others attempt to argue both that self-knowledge’s peculiarity must be explained in non-epistemic terms and that our self-ascriptions express knowledge. Regarding the former, some might object that “self-knowledge” a misnomer (see Coliva, 2016, p. 163). And, as such, those “minimally epistemic” views cannot compete with other theories of self-knowledge. The latter face the challenge of showing that our avowals express knowledge—widely understood as justified second-order belief—even though their peculiar status derives from some non-epistemic aspect.
Those understanding self-knowledge’s distinctiveness in terms of the metaphysics of mental states minimally agree that we have privileged knowledge of our own mental states owing to their nature. Their answer to the question of how I know my mind seems simple (at least at first glance): it is by being in that mental state. The main challenge for this kind of view is to show that avowals express knowledge. If being in a mental state necessarily comes with one’s awareness of that state, then it is unclear how that awareness can count as some form of knowledge. As Paul Boghossian says (1989, p. 5), self-knowledge would be based on nothing; it would not involve any effort on the subject’s part.
To compete with other theories of self-knowledge, we must say more about why the kind of awareness that comes with being in the mental state can count as self-knowledge. As I shall discuss in Section 1.1, proponents of the transparency view of self-knowledge attempt to do this. They also offer an answer to the question of how I know my own mental states: in virtue of the nature of our mental states, I am entitled to answer a mind-directed question by attending to the world. As we shall see, the transparency view (specifically a certain interpretation thereof) can give substance to the intuition that being in a certain mental state is one thing and knowing that mental state is another. For now, it suffices to say that the transparency claim (that I can answer the mind-directed question by attending to the world) is grounded in claims about the nature of the mental state in question.11
In a nutshell, the transparency view that I defend here (which I believe comes closest to Moran’s view) is difficult to locate on this map.12 It does not fall clearly on either the epistemic or the non-epistemic side. That said, it clearly entails that self-knowledge is not acquired in the way that other forms of knowledge are. And spelling out this other way goes beyond epistemology. Our self-knowledge is peculiar because we are agents and see ourselves as such rather than because we are experts in the way that Freud’s doctor is.
Having briefly introduced the topic of self-knowledge, I shall now provide a preliminary sketch of the transparency view and its most influential versions. In doing so, I hope to further specify this book’s objective.
2 Looking In, Looking Out
To get closer to the core of so-called transparency views, it is common to begin by quoting a (notorious) passage from Gareth Evan’s The Varieties of Reference:
[I]n making a self-ascription of belief, one’s eyes are, so to speak, or occasionally literally, directed outward—upon the world. If someone asks me ‘Do you think there is going to be a third world war?’ I must attend, in answering him, to precisely the same outward phenomena, as I would attend to if I were answering the question ‘Will there be a third world war?’ I get myself in a position to answer the question whether I believe that p by putting into operation whatever procedure I have for answering the question whether p. (1982, p. 225)
Evans is incisively pointing out that we normally attend to the world around us, rather than ourselves, when self-ascribing a belief. We do not observe ourselves or look for evidence of our own mental states’ existence. Instead, we look out for non-mental facts.
Evans’s observations are notable for being concurrently simple and substantive. They elucidate privileged self-knowledge by highlighting its commonplace nature and appealing to our intuitions about beliefs and self-ascriptions. Moreover, as many self-knowledge theorists have noticed, his observations offer a potential criterion for distinguishing between avowals and other forms of mental state ascriptions. In its most basic form, the transparency theorist claims that I can know whether I believe that p by considering (nothing but) p itself (see Gallois, 1996; Moran, 2001; Byrne, 2005; Boyle, 2011; Fernández, 2013). When answering the world-directed question by settling my mind about p, I can, thereby, also answer the mind-directed question about my belief that p. Doing so eliminates the need for a mind-directed observation or interpretation.
Various versions of the transparency view have been put forward, although they all adhere to Evan’s key idea: I need not attend to my own mind to determine whether I believe that p. Although the relevant authors agree that the mind-directed question is transparent to the world-directed question, they disagree about several key aspects. However, before moving on to examine the differences between these viewpoints, I should say a few things about the transparency view’s overall advantages. Four such advantages stand out.
The first advantage of such a view is that one does not have to postulate a special faculty of inner sense or internal scanning. The transparency view takes most of self-knowledge’s mysteriousness away without depriving it of its peculiarity. To use Alex Byrne’s words, this means that the transparency view does not require a capacity or mechanism over and above our general epistemic capacities or normal capacity for rationality (2005, p. 99, 2012, p. 207). All transparency views are economical in that they do not posit some inner eye. Nonetheless, we shall have to see whether some are “extravagant,” rather than “economical,” in a different way (e.g., by holding rather extravagant assumptions about the nature of our mental states).13
The second advantage is that the transparency view can explain self-knowledge’s peculiarities. To decide whether p, I might have to consider evidence that helps me settle my mind on p. But my observations are about p and do not have my mind as their object. So, the claim is that coming to know my beliefs does not require self-observation. This means that I can come to know my beliefs immediately by settling whether p. (As will be shown, this does not apply to all versions of the transparency view, however.) Much more must be said about (a) the “transition” from answering the world-directed question to answering the mind-directed question and (b) how self-knowledge can still be immediate if there is a need for this transition. Nonetheless, for now, note only that it is worth looking into the transparency view’s explanation of self-knowledge’s distinctiveness.
The third advantage lies in the transparency view’s intuitive appeal. It is simply what it seems like or what we seem to do when asked what we believe. When asked what I believe, it seems very natural to attend to the content of my beliefs. For example, if someone asks me whether I believe that it is raining, I normally look through the window or watch the weather forecast before coming to the conclusion that it is raining and then self-ascribing the belief that it is raining.14 It will have to be seen whether it is only intuitively appealing with regards to a small group of mental states (i.e., conscious, occurrent beliefs) or whether the same applies to phenomenal states.
The fourth advantage is that the transparency view can explain why self-knowledge is something I acquire while accounting for its distinctiveness in terms of a constitutive relation between my belief and my being aware of my belief. As we shall see, some versions of the transparency view can more clearly account for the intuition that self-knowledge involves some sort of cognitive achievement than others. The different views seem to agree that it can take some effort to make up one’s mind about the world-directed question and that one must then move from answering the one to answering the other question. But we must still discuss the extent to which the subject might acquire self-knowledge and whether the explanation for this acquisition is compatible with the aforementioned advantages.
Transparency views, put broadly, fall into two camps. Following Brie Gertler’s above-mentioned distinction, Alex Byrne’s inferentialist view (2018) is paradigmatic of an epistemic approach, whereas Richard Moran’s (2001) and Matthew Boyle’s (2011, 2024) views exemplify non-epistemic approaches. As mentioned, Moran’s view seems to encompass both epistemic and non-epistemic aspects. However, when it comes to the question of what renders self-knowledge distinctive, we are looking at the non-epistemic aspect.
Moran (2001) has developed Evans’s observations into a theory of self-knowledge. He argues that we usually answer the question “Do I believe that p?” by making up our minds, determining whether to believe that p, or answering the question of whether p is true. On Moran’s view, any account of self-knowledge that understands self-knowledge in terms of inner sense, inference, or observational evidence fails to account for self-knowledge’s immediacy and authority. Instead of explaining self-knowledge’s distinctiveness in these terms, Moran draws on its relevance for rational agency and criticizes detectivist theories of self-knowledge for failing to acknowledge the specific results self-consciousness causes for its object. At the heart of his view is the idea that the capability to make up my own mind about what I intend, believe, or want explains the distinctive character of first-personal self-knowledge. Our mental states stand in a special relation to our reflective, deliberative standpoint, which can be characterized by the following features: we take our attitudes to be open to change and are aware that they can be shaped through deliberation. Our knowledge of our attitudes is not based on observation, and our attitudes are transparent to us in that we can gain knowledge of them by looking directly at their objects.
However, this is not always the case. From time to time, I may turn into a spectator of myself: I discover or detect whether I believe this or that. Moran calls this the “theoretical” stance toward oneself, but the knowledge at issue here is akin to external knowledge—knowledge I can have about others and the world. The latter does not explain the first-person authority with which we ascribe our beliefs to ourselves. From the deliberative stance I must take into consideration reasons for and against believing that p if I want to address the question of whether or not I believe that p.15 But if I am addressing the same question from a theoretical stance, then I reason for and against believing that someone believes that p. Thus, the capability to avow one’s mental states (e.g., to self-ascribe a belief by attending to p) makes “the difference between genuine first-person awareness and a purely theoretical or attributional knowledge of one’s own states” (Moran, 2001, p. 107).
Importantly, Moran understands transparency as a condition of agency, identification, and rationality. But he also denies that self-knowledge is simply a conceptual truth and not something we might want to achieve. I can achieve knowledge of my belief that p, but by committing myself and thereby making up my mind (literally), by making it the case that I believe that p, rather than detecting or finding out that I believe p.16
Matthew Boyle (2011) has added the following claim: the move “from believing p to reflectively judging (i.e., consciously thinking): I believe p” (2011b, p. 227) can be understood as “coming to explicit acknowledgment of a condition of which one is already tacitly aware” (Boyle, 2011b, p. 227). Boyle further claims that
the very same actualization of my cognitive powers that is my believing p is, under another aspect, my tacitly knowing that I believe p. Hence, to pass from believing p to judging I believe p, all I need to do is reflect—i.e., attend to and articulate what I already know. (2011b, p. 229)
This is why Boyle refers to his view as “reflectivist.” Although he largely agrees with what Moran says, Boyle also goes beyond Moran’s view by introducing the notion of tacit knowledge and emphasizing the process of bringing what one is tacitly already aware of to consciousness.17
Although transparency is also at the core of Byrne’s view, his approach differs substantially from Moran’s and Boyle’s. Byrne’s view can be summed up as follows: we can acquire self-knowledge by making a reliable “inference from world to mind” (Byrne, 2011, p. 203). This is an inference from the premise that p to the belief that p. According to Byrne, one
finds out that one believes that it’s raining by determining that it’s raining: knowledge that one has this belief […] rests on perceptual evidence about the weather, not on perceptual evidence of one’s behavior or anything mental. That is, one reasons from evidence that it’s raining, to the conclusion that one believes that it’s raining. (2005, p. 93)
If I follow the epistemic rule “If p, believe that you believe that p,” then I acquire knowledge of my belief by considering evidence that is relevant to the formation of the belief in question. Byrne believes that this procedure’s strongly self-verifying character is something special: to follow the rule, I must first “recognize that p” and “recognizing that p is inter alia coming to believe that p” (Byrne, 2005, p. 96). My recognition that p epistemically grounds my belief that I believe that p.
Here, the special authority of self-ascriptions derives from the fact that “beliefs about one’s mental states acquired through the usual route are more likely to amount to knowledge than beliefs about others’ mental states (and, more generally, beliefs about one’s environment)” (Byrne, 2005, p. 80). In short, the self-ascription “I believe that p” expresses knowledge because the inference from p to “I believe that p” results in a justified and reliable self-ascription. Inferring in line with the transparency method is reasonable because it (usually) leads to a true and justified belief.
As Byrne rightly notes, views such as the one Moran defends differ from his view in that they take transparency to be related to self-constitution, while, for Byrne, transparency is a way of self-detection (2005, p. 83). Consequently, the main challenge that proponents of “non-epistemic,” agentialist, and/or reflectivist transparency views face is showing how their views can count as theories of self-knowledge.
My aim so far has been to introduce the transparency view. In Chapter 1, I intend to defend a specific version thereof. By looking more closely at how transparency between the two pertinent questions is related to first-personal self-knowledge, I shall argue that only a specific version of the transparency view can preserve its above-mentioned advantages. This view comes closest to the view Moran and Boyle defend. I shall call this view the Phenomenal Reflectivist Transparency View.18
3 Preview
In writing this book, I have two primary goals:
1. Defend a specific version of the transparency view.
2. Make a case that the transparency view, or a certain version of it, can be extended to mental states other than belief.
Accordingly, my goal in Chapter 1 is to defend the transparency view against two major objections:
1. The view explains away what it wants to explain, which is self- knowledge.
2. The objection of scope: the objection that I can come to know only my beliefs, and not my other mental states, in the way that the transparency view puts it.
In Chapter 1.1, I attend to the first objection and answer the question about whether the transparency view can count as a positive theory of self-knowledge. I shall ask how the different views can solve the problem of the two subject matters (which constitutes the first horn of the transparency dilemma): how can reflection about the world possibly settle a question about my mind? How can we capture the intuitively appealing idea that we attend outwards when asked what we believe while arguing that evidence about p can come apart from what we believe about p? An adequate answer should either avoid presupposing self-knowledge (the second horn of the transparency dilemma) or show why it is unproblematic to embrace the second horn. Roughly, my argument will be that the kind of awareness presupposed is not awareness of my judgment as a mental fact about myself: what I am aware of is p as held to be true.
I shall also discuss whether, and in what sense, the ‘non-epistemic’ transparency view can do the following given its answer to the dilemma: explain the distinctiveness of self-knowledge in non-epistemic terms while still holding onto self-knowledge’s epistemic authority. On the view I defend, mental states are understood as ways of relating to the world. And my attending to the world to answer the mind-directed question reveals awareness of both the world and my way of relating to the world. When expressing a belief, I do not only express the manifest content of my belief (e.g., that eating vegetables is healthy). I also express my endorsement of the manifest content (e.g., my belief that eating vegetables is healthy), thereby reporting on some mental fact about myself and answering the mind-directed question. Importantly, this need not imply that my belief is an antecedent mental fact I detect via inference or observation. I have a certain awareness of the world (an awareness of something as true in the case of belief) that self-ascribing my belief makes explicit. This is not only crucial for my proposed defense of the transparency view against the scope objection in the chapters to come. It is also crucial for defending it against the objection that the transparency view explains away what it wants to explain, which is self-knowledge.
In Chapter 1.2, I address the scope objection and pave the way for my extension of the transparency view (Chapters 2–4). Section 1.2.1 (All Beliefs?) is concerned with the objection that the transparency view cannot account for dispositional beliefs, which those raising this objection consider to be the central class of beliefs. Section 1.2.2 (Beyond Beliefs?) is focused on the objection that the transparency view (as defended by Moran and Boyle) is limited to a small group of mental states. To extend the transparency view in the following chapters, it will help to first look closer at this objection, how scholars have responded to it, and what such an extension could amount to.
I shall also argue that there are good prima facie reasons to favor a unified account over a hybrid or pluralist account of self-knowledge (according to which we have a different kind, or different kinds, of knowledge about our phenomenal mental states and sensations). Unlike a hybrid or pluralist approach, a uniform approach of self-knowledge can, I contend, account for (a) phenomenal states (e.g., emotions) that are best understood as unified experiences rather than combinations of feelings plus intentionality and (b) continuity between our mental states (e.g., between emotional and physical pain).
If the transparency view was limited to beliefs, then one would have to explain how the other kind of self-knowledge is related to the transparent self-knowledge of belief. And one would have to do so in a way that both avoids self-fragmentation (a person might end up having transparent self-knowledge of her beliefs and belief-like states but not of her desires, emotions, and sensations. Alternatively, she might know part of her emotions transparently and another part of her emotion through inner sense or observation) and might hold onto the epistemic asymmetry assumption (i.e., all knowledge of my present mental states is distinctive in that it differs essentially from my knowing the external world). Accordingly, the line of argument I pursue in this book will not involve simply showing that mental states other than belief are belief-like enough for the transparency view to apply. Doing so would ignore the phenomenological differences between our different mental states. It would also fail to explain how I can transparently know the kind of mental state I am in by only attending to its object.
The objects of our different mental states might not differ (e.g., I can believe that flying is dangerous, I can fear flying, or I can desire flying). This means that my awareness of the object alone, so to say, cannot justify my self-ascribing a certain kind of mental state. Solving this challenge of individuation is inevitable for any transparency view of self-knowledge; so is adequately addressing cases of recalcitrant emotions or desires. Before addressing these issues in more detail and in relation to the nature of different mental states, I shall list some preliminary assumptions in Section 1.3 (Beyond Beliefs: Intentional Modes) regarding the intentional structure of our mental states, the location of their phenomenal character, and the relationship between mode and content. Doing so is relevant for both defending the phenomenal reflectivist transparency view and for extending it.
Chapters 2–4 contain the heart of this book (Chapter 2: Transparent Self-Knowledge of Desire; Chapter 3: Transparent Self-Knowledge of Emotion; Chapter 4: Transparent Self-Knowledge of Pain). These chapters are both defensive and constructive in that they challenge the scope objection by extending the transparency view beyond beliefs. I shall defend the transparency claim (the claim that we can tell which state we are in by directly attending to the objects of the states in question). And I shall do so in a way that (a) does justice to the differences between our mental states and (b) avoids the rationalism or intellectualism that some critics think the transparency view entails. In this regard, my main concern will be distinguishing and comparing the ways in which we know our different mental states.
I shall pursue roughly the same strategy in Chapters 2–4. Each one starts with a question about what the mental state in question is and then asks how this is compatible with the transparency view. In so doing, I shall also argue that the inferentialist’s version of the transparency view cannot account for the distinct nature of the mental states in question. In particular, it cannot do justice to both the world-directedness and the distinctive character of the different states while maintaining the transparency view’s advantages.
As summed up in the conclusion, I hope to have outlined a transparency view that has the capacity to be a unified theory of self-knowledge while avoiding harsh rationalism and accounting for the distinctive character of different mental states.
In Self-Knowledge for Humans, Quassim Cassam complains that the academic discourse focuses on trivial self-knowledge, such as the knowledge that I believe that I am wearing socks (2014, p. 28). He claims that there are far more interesting questions around what he calls “substantial self-knowledge,” including knowledge of one’s own character, values, abilities, aptitudes, and emotions, as well as knowledge of what makes one happy.
In Notes Upon a Case of Obsessional Neurosis (1909), Freud discusses the case of a patient nicknamed “Rat Man” on account of his obsessive fear of rats. Freud interpreted this fear as resulting from Rat Man’s repressed emotions toward his father. As I am aware of the many controversies in Freud’s psychoanalysis, including the lack of empirical evidence for his interpretations of cases such as these, I do not intend to subscribe to it here by using these examples. But I think that Freud, as becomes especially apparent in the above quotation, might have noted an important difference between two ways of knowing one’s own mental states in mind that has similarities with the difference between first- and third-personal self-knowledge as discussed in the current philosophical literature on self-knowledge.
The insight that the question of self-knowledge is intertwined with the metaphysics of our minds has led to different views on the relation between the epistemology of self-knowledge and the metaphysics of our mental states. I shall discuss this further shortly.
There have been recent worries that approaching the problem of self-knowledge in terms of self–other asymmetry frames the former as a discrete epistemological problem, thereby dissociating it from metaphysical claims about the nature of our minds (Boyle, 2024, p. 3). I shall, nonetheless, follow the common practice of comparing self-knowledge and other-knowledge. This is because I do not think that it dissociates the problem of self-knowledge from the metaphysics of our minds. As will become clear, I also believe that the question of self-knowledge is an epistemological one (not only in the basic sense that it belongs to the study of knowledge but also in the sense that it is essentially a question about the process of acquiring knowledge).
Some philosophers consider the asymmetry to be more “profound” (Boghossian 1998, p. 151) or “categorical” (Moran 2001, p. xxxi) than others (Cassam 2015).
Gilbert Ryle (1949) argued that the apparent asymmetry between self-knowledge and knowledge of others is an illusion. He believed that self-knowledge is based on evidence, and that the difference between self-knowledge and other-knowledge is only a matter of degree.
On David Armstrong’s (1968) inner-sense view, some internal self-scanning process that monitors one’s brain activities is specialized to give one access to one’s own mind. Here, self-knowledge gained through inner observation involves causal processes similar to those involved in outer perception.
Introspectivism (entailing some mysterious infallible inner eye or internal tracking mechanism) is one such epistemic view. Inferentialism is another. On one version of inferentialism, inferences from psychological and/or behavioral evidence are the sources of self-knowledge (Lawlor, 2009; Cassam, 1996, 2014). On another version, one draws inferences in accordance with a certain epistemic rule to acquire particularly reliable self-knowledge (Byrne, 2005, 2011). I shall return to Lawlor’s and Cassam’s view in Chapter 2.
On the neo-expressivist view, the distinctiveness of mental state self-ascriptions lies in the fact that they directly express, rather than descriptively report, our mental states. Different objections have been raised against neo-expressivism. Most of these relate to the attempt to offer a non-deflationary, non-epistemic account of self-knowledge. Very roughly, the problem involves showing how self-ascriptions are more than blind expressions while maintaining that their distinctiveness is attributable to their expressive nature. A discussion of the neo-expressivist view (Bar-On, 2004; Finkelstein, 2003) is beyond the scope of this book. Essentially, I agree with Boyle (2010, p. 16) that the expressive act can ground self-knowledge or (more generally) first-person authority only if there is an explanation for how this act is already self-conscious (2010, p. 16).
Crispin Wright (2001, p. 310) and Paul Boghossian (1989, p. 5) argue that because self-knowledge is based on nothing, it cannot be an epistemic achievement. The distinctiveness of our first-person self-ascriptions is a conceptual necessity and, as such, is not an instance of self-knowledge.
Moran (2001) and Boyle (2024) aim to reorient our understanding of self-knowledge in a way that allows us to go beyond traditional epistemological questions about the nature of self-knowledge. We can then consider the human mind’s wider context and peoples’ agency and rationality. Nonetheless, we should not overlook Moran’s goal to “explain and vindicate avowal as a privileged form of knowledge of oneself” (2001, p. 134). He thinks that self-knowledge’s distinctiveness (an epistemic dimension) stems from our agency (which is non-epistemic). However, the relationship between these dimensions is not entirely clear (I discuss this issue further in Section 1.1.3).
See also Jongepier and Strijbos (2015, p. 132).
I shall also discuss in Section 1.1 whether being economical is still an advantage if the transparency view implies extravagant assumptions about the nature of mental states. Boyle (2011b, 2024), for example, claims that believing involves some kind of tacit awareness of believing.
Mostly, there is no need for any explicit deliberation. In these cases, I am still (implicitly) holding the contents of my beliefs to be true (but more on this later).
Insofar as a subject attends to her beliefs from a deliberative stance, she asks what to believe.
I discuss this possibility in more detail in Section 1.1. For now, note that this view can count as a ‘non-epistemic’ version of the transparency view, one that attempts to account for self-knowledge understood as cognitive achievement.
At times, it seems that Boyle does not think that we need a story about how we gain self-knowledge. But, at other times, he claims that our self-ascriptions are justified owing to our reflection on the world. This is because we are already implicitly aware of our mental states in believing. In Chapter 1.1, I hope to shed more light on Boyle’s view.
Boyle calls his view “reflectivist” (Boyle, 2011, 2024). My view largely aligns with Boyle’s approach, with the adjective ‘phenomenal’ serving to emphasize a modification I introduce.