What do you think?
Rate this book
[(Knowing Our Own Minds)] [Author: Crispin Wright] published on
Knowledge of one's own sensations, desires, intentions, thoughts, beliefs, and other attitudes is characteristically different from other kinds of it has greater immediacy, authority, and salience. This volume offers a powerful and comprehensive look at current work on this topic, featuring closely interlinked essays by leading figures in the field that examine philosophical questions raised by the distinctive character of self-knowledge, relating it to knowledge of other minds, to rationality and agency, externalist theories of psychological content, and knowledge of language.
- GenresPhilosophy
Unknown Binding
First published December 17, 1998
2 people are currently reading
40 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 of 1 review
March 19, 2025
What is the nature of our apparent distinctive first-person authority over knowing some of our mental states consists in? There are different ways of taking this. Some essays focus on the epistemological questions of whether we indeed have this authority, and if so, whether it is justified. Others focus on metaphysics of mind-ey questions of what processes do we psychologically/mentally go through in obtaining this special form of knowledge. The topic is relevant to debates on issues such as self-deception and implicit biases, but I don’t see it brought up so much.
I was struck by the exchange between Christopher Peacocke and Michael Martin here. In “Conscious attitudes, attention, and self-knowledge” Peacocke offers a picture of how we come to know our beliefs, one which vindicates that it is epistemically well-grounded. In brief, his argumentative strategy is to explain what the constitutive conditions of possessing a belief amount to and to show that these conditions are normally satisfied, regarding our possessing a standing second-order belief, once we possess a first-order belief.
He points out that there is a tendency in offering such pictures to turn off into one of two extremes. On one end, there are views which make out such self-knowledge of our beliefs to be like perceptual judgment. Instead of looking at the environment and forming a perceptual belief, we look inwards and form a belief about our belief. These views make sense of the legitimacy of this knowledge in terms of reliability; the knowledge is reliable, as a product of purely causal mechanisms, as perceptual knowledge is. On the other extreme, there are views which make out such self-knowledge to be a product of drawing inferences and reasoning. On such views, this knowledge is legitimate not because it is merely reliable, but because it is also achieved through well-grounded reasoning and practices of justification.
Peacocke avoids these two extremes. On his view, we need not consciously or deliberately reason about our mental states in order to arrive at belief about our own beliefs. Such second-order belief even need not be conscious or occurrent in order for us to possess it. Instead, this second-order belief can be “standing”; that is, we can readily articulate it once we are prompted to do so, and our behavior and reasoning reflects that we possess this second-order belief. Peacocke argues that such standing second-order belief is epistemically grounded in the first-order belief, which it attributes to us, by virtue of the fact that this first-order belief serves as a reason for us to have formed that second-order belief. This “serving as a reason” relation, again, need not depend upon our deliberately or ocurrently reasoning about it.
An upshot is that his view gives a way of understanding why whether or not a belief is conscious and occurrent need not correspond to whether that belief is “inferentially integrated” with the rest of the mind. Normally, a belief is prone to influencing our reasoning at large if we’ve consciously and occurrently thought about this belief and attributed it to ourselves. But this need not be the case. Some unconscious beliefs have great power over us (think of cases of people who have repressed their low self-confidence; while they’d report that they’re great, their bearing doesn’t reflect that), and some conscious/occurrent beliefs don’t have much influence (Peacocke’s example, taken from Dennett, is a father who self-attributes the belief that his son is a great artist, but his day-to-day attitudes don’t quite align with that). Peacocke’s view illuminates this insofar as it offers a way for self-attributed or higher-order beliefs to be standing and unconscious.
Martin’s essay “An eye directed outwards” critically assesses Peacocke’s position. Martin provides the useful distinction between the issue at hand and a related one, that which surrounds later Wittgenstein’s proposal that knowledge of our own inner states differs in nature from that of other people’s states. Regarding the issue of hand, we are contrasting knowledge of our own beliefs gained through non-inferential, seemingly automatic means, on the one hand, and knowledge of our own beliefs gained through inferential or “inner observation”-based means, on the other hand. It’s a first-person issue through and through. So the difference between these two varieties of first-person self-knowledge can’t be explained by appeal to metaphysical differences; that isn’t guaranteed. Peacocke is right to look at epistemological considerations to address this issue.
Martin points out that while Peacocke claims that we form, often unconsciously, second-order beliefs just by looking out onto the world and gleaning first-order beliefs, he hasn’t explained exactly how this happens, or why directing our attention outwards results in such self-knowledge. In other words, while his proposal is plausible, that we have second-order beliefs just in having first-order beliefs, Peacocke hasn’t showed how this occurs.
Moreover, this proposal faces conceptual issues which Peacocke hasn’t considered. Martin points out that some of Peacocke’s reasoning dangerously relies upon phenomenological considerations. Self-knowledge of our beliefs can arise seemingly spontaneously as perceptual knowledge does. But phenomenology is under-determined regarding the nature of things. Namely, there is a certain distinction that Martin shows is central to the issues at hand, and it isn’t readily picked up via phenomenology, and rather relies upon conceptual consideration. There’s a distinction between something being an object of our attention, on the one hand, and something occupying our attention, on the other hand. For example, when the content of a thought is an object of our attention, we know that content; but when that content simply occupies our attention, all we can say is that we’re implicitly aware of having thought, but this need not imply that we could report on what the contents of that thought are. This distinction is particularly relevant when we talk about belief. In contrast, when we talk about perception, it does not matter as much because the notion of perception suggests the notion of relationality to the content of perception: when a perceptual experience occupies our attention, we are standing in relation to the perceptual object in the world, and so by necessity there’s an object of our attention. When thought and imagination enter the picture, we’re no longer in relation to their objects; there is no object of attention (in the relational sense) that having a thought or imagining guarantees, because they don’t have objects in the first place.
When it comes to Peacocke’s claim that we spontaneously inherit reasons, from looking at the world around us, which amount to the formation of a first-order belief and the guarantee that they can serve as reasons for second-order belief that we have that first-order belief, this distinction is critical. Looking at the world, as a perceptual affair, involves this relationality. But in forming beliefs about our beliefs, which can sometimes consist in primarily or only thought, we don’t have this relationality; it’s not like there’s a landscape of our beliefs set in stone and in a determinate form already there, and we respond to that. Martin thinks that Peacocke implicitly understands this and would hold that having thought consists in a determination of our attention, rather than there being thought as something independent of our attention and to which we direct our attention.
This distinction implies that more is required for a creature’s capacity to form second-order beliefs than the resources which suffice for forming perceptual knowledge. It shouldn’t be required that a creature possess the concept of beliefs; it’s natural to think that pre-linguistic humans can have awareness of what they believe. It also shouldn’t be that having a first-order belief itself constitutes a second-order belief regarding it; the two can come apart, for it is intuitive to think that sometimes knowledge of our beliefs require indirect routes, like reliance on the testimony of people who know us. But then what is the picture here? Martin points out that Peacocke doesn’t offer one. Instead, Peacocke only proposes, schematically, that the reasons for first-order belief are inherited as reasons for that belief to serve as a reason for a second-order belief in such a way that the essential conditions for the formation of that second-order belief are satisfied, and so it makes sense to say that we possess that second-order belief, in at least a standing, not occurrent, form.
Here’s another problem with Peacocke’s proposal. We don’t always look at the world in order to form first-order belief. Some of our first-order beliefs concern theoretical issues. Martin asks the reader to think about the example of thinking through a difficult philosophical problem. Here, we might not have a strong first-order belief about the matter (e.g., I don’t have a conclusive belief about the best way of understanding belief under psychological, rather than epistemological, contexts). Then, in order to know what we believe at the first-order, we might rely upon an inner feeling of conviction; we can propositionally imagine alternative positions and sense which one of them we feel most convinced by. In this case, we ought not expect that the reasons for holding that first-order belief transfer up to the level of the second-order belief; we don’t hold that second-order belief for the same reasons, for the new element of a feeling of conviction is part of the reason-base here.
In such cases, in what sense has the first-order belief served as a reason for holding the second-order belief? This is a problem for Peacocke’s view. Cases of self-deception, like the examples I gave above, can be understood in these terms as well. Such cases also bring out how it’s easy to get things less than right in theoretical cases. It’s very much possible that when we reflect upon a theoretical proposition in addressing a philosophical problem and we feel a strong conviction that, in fact, this conviction is responding to some other proposition which is related but distinct from this current one. In these cases, we can be less than trustworthy in our first-person beliefs about our beliefs, and this fact brings out just how tricky it is to demarcate the set of second-order beliefs, arrived at first-personally, which are trustworthy.
Martin points out that all we can say here is that there’s some “primitive rationally compelling transition” here, located between the first- and the second-order belief, and it should not be reduced to the perceptual kind. But given all of the problems which Peacocke’s view faces, Martin points out that given the current status of the debate, it’s not obvious or compelling that self-knowledge of our beliefs shouldn’t be reduced to that perceptual kind. Martin thinks Peacocke’s contribution to the debate is to make explicit that the proposal is that there’s such a primitive rationally compelling transition which is irreducible to the perceptual kind, and it leaves it to future thinkers to spell out what it consists in and to defend its existence.
Okie, so there’s a very rough overview of those two essays. Now, here are some of my ramblings in response to them. When I found myself first drawn to this topic of self-knowledge, I wasn’t quite sure why; on the surface, it doesn’t have much relation to what I’ve been focusing on (e.g., emotion regulation, and most locally, repressed emotions). Now, I see some potential connections. First, this debate shows that when we define what it means for some mental phenomenon to be conscious or unconscious, there are ways of doing it which have nothing to do with whether we are deliberately or occurrently transacting with it. Instead, it can be a matter of whether a “standing belief” exists about our having that mental phenomenon, where that mental state can be minimally understood in terms of that mental phenomenon properly figuring in as a personal reason for having that state such that we could report on it if we were appropriately prompted. It can also be a matter of how “inferentially integrated” that mental phenomenon is, which maybe can be understood as an independent consideration of that of whether we have a standing belief about it.
All of this is on top of Block’s famous distinction between access- and phenomenal-consciousness. What do Peacocke and Martin’s topic of study in these essays have to do with Block’s distinction? They are orthogonal; Peacocke’s account holds that in order to have a standing second-order belief (which could be glossed as access-conscious) one needs to be experiencing the world about (which could be glossed as phenomenally conscious) which one has the first-order belief which that second-order belief attributes to oneself. Other connections could be drawn, but they are independent matters. Block is focused on bringing out how relating to the mind phenomenally is quite different from relating to the mind via offering functional-level descriptions of its parts, whereas Peacocke and Martin in these essays are focused on addressing certain epistemological problems in such a way that it is compatible with their accounts to reject that those two types of description of the mind are basic, distinct, or exhaustive.
So to return to my main thought: I don’t think I should proceed in my writing by presupposing some sense of what it means for a mental phenomenon to be conscious or unconscious. That issue is just too complicated. Instead, I could proceed by talking about the trustworthiness of self-reports of the mental phenomena one is undergoing or possesses, where this trustworthiness could be understood in terms of either reliability or justifiability. In that case, when we speak of repressed emotions and desires, these could be understood in terms of that the subject has difficulty forming trustworthy self-reports about these, and we can thereby dodge the issue of whether the subject has “experienced” or “phenomenally encountered” the emotion or desire under consideration, or whether it is “accessible” in the sense that it can informally guide or control rational or behavioral activity or not.
Here’s another thought I have in response to the issues posed by Peacocke and Martin. I wonder whether the “primitive rationally compelling transition” which links first- and second-order beliefs must be open to variation in degree regarding how trustworthy and justifiable the resultant report on one’s belief is. Once the first-order belief under consideration is not a perceptual belief, and is rather partially constituted by imagination/thought qua a determination of one’s attention, it is possible that it is also stands as partially a determination of one’s interests and values. For example, when I think about philosophical issues, I am often interested in one thesis over another holding true, namely the one which would vindicate a position I’ve been married to for a while now. This is just a fancy way of saying that beliefs can be motivated or driven by desire. Maybe there’s one kind of primitive rationally compelling transition which defines cases in which our self-knowledge of our beliefs is transparent, automatic, or infallible, but the direction of this transition is highly reactive to desire/motivation/emotion. In some cases, this won’t be taken as impacting the validity of the knowledge because the starting first-order belief in the first place wasn’t totally formed, and so this “transition” is partially constitutive of the second-order belief; so questions of validity aren’t only essential, but it is also irreducibly partially a practical/ethical matter of what beliefs one self-constitutes. But in other cases, it will be taken as impacting the validity, namely those under which the starting rest-order belief was already well-formed.
I guess the authors might want to distinguish between knowledge of currently existent beliefs, on the one hand, and our making up our minds or changing our beliefs (to which coming to know our beliefs can sometimes amount to) on the other hand. My concern raised above confuses these two issues. One could rather say that when an extant belief is very general or non-committal, relative to beliefs at a more fine-grained level, which is the one that we're ultimately aiming to arrive at, it more effectively eludes our knowing it, or it will be especially responsible and malleable to our act of attempting to know it. So we shouldn't say that the primitive rationally compelling transition at hand admits of degrees in the way I suggested above; instead, when this kind of transition occurs, depending upon the level-of-grain of the starting belief involved, it can sometimes require that other sorts of cognitive activity, and even the subject's agency and desires, come into the picture, to help complete the transition. These musings indicate, however, that it's tricky; perhaps trying to specify the fine-grained processes involved in self-knowledge is futile, and instead we should stay at a rougher grain of analysis.
Reading these essays also provokes reflections on methodology in philosophy. Something seductive about the Peacocke-Martin exchange is that their approach gives the feeling of using a high-precision microscope, which focuses in on the subject of our knowledge of our beliefs. In contrast, other thinkers I’ve encountered on this subject matter can feel blunt, only skimming the surface (e.g., Krista Lawlor’s “Knowing what one wants”). I think this difference in feeling is based upon the sorts of considerations these different thinkers use in proceeding in argument. Lawlor’s paper consists in reflections upon a personal example. In contrast, Peacocke’s and Martin’s papers consist in raising questions of the commitments tied up to an intuitive claim, surveying the landscape of competing claims, and comparing their respective commitments. I think these different approaches each have an important role in the collective project of advancing our understanding of the issues at hand. Here’s a way of articulating the roles of these approaches. It’s important to look at phenomenology and personal examples for expanding the hypothesis space and for argumentation under the form of finding counterexamples of extant claims, but it’s critical to not draw conclusions between competing hypotheses without focusing on claims with respect to where they are nestled within implied conceptual networks.
I was struck by the exchange between Christopher Peacocke and Michael Martin here. In “Conscious attitudes, attention, and self-knowledge” Peacocke offers a picture of how we come to know our beliefs, one which vindicates that it is epistemically well-grounded. In brief, his argumentative strategy is to explain what the constitutive conditions of possessing a belief amount to and to show that these conditions are normally satisfied, regarding our possessing a standing second-order belief, once we possess a first-order belief.
He points out that there is a tendency in offering such pictures to turn off into one of two extremes. On one end, there are views which make out such self-knowledge of our beliefs to be like perceptual judgment. Instead of looking at the environment and forming a perceptual belief, we look inwards and form a belief about our belief. These views make sense of the legitimacy of this knowledge in terms of reliability; the knowledge is reliable, as a product of purely causal mechanisms, as perceptual knowledge is. On the other extreme, there are views which make out such self-knowledge to be a product of drawing inferences and reasoning. On such views, this knowledge is legitimate not because it is merely reliable, but because it is also achieved through well-grounded reasoning and practices of justification.
Peacocke avoids these two extremes. On his view, we need not consciously or deliberately reason about our mental states in order to arrive at belief about our own beliefs. Such second-order belief even need not be conscious or occurrent in order for us to possess it. Instead, this second-order belief can be “standing”; that is, we can readily articulate it once we are prompted to do so, and our behavior and reasoning reflects that we possess this second-order belief. Peacocke argues that such standing second-order belief is epistemically grounded in the first-order belief, which it attributes to us, by virtue of the fact that this first-order belief serves as a reason for us to have formed that second-order belief. This “serving as a reason” relation, again, need not depend upon our deliberately or ocurrently reasoning about it.
An upshot is that his view gives a way of understanding why whether or not a belief is conscious and occurrent need not correspond to whether that belief is “inferentially integrated” with the rest of the mind. Normally, a belief is prone to influencing our reasoning at large if we’ve consciously and occurrently thought about this belief and attributed it to ourselves. But this need not be the case. Some unconscious beliefs have great power over us (think of cases of people who have repressed their low self-confidence; while they’d report that they’re great, their bearing doesn’t reflect that), and some conscious/occurrent beliefs don’t have much influence (Peacocke’s example, taken from Dennett, is a father who self-attributes the belief that his son is a great artist, but his day-to-day attitudes don’t quite align with that). Peacocke’s view illuminates this insofar as it offers a way for self-attributed or higher-order beliefs to be standing and unconscious.
Martin’s essay “An eye directed outwards” critically assesses Peacocke’s position. Martin provides the useful distinction between the issue at hand and a related one, that which surrounds later Wittgenstein’s proposal that knowledge of our own inner states differs in nature from that of other people’s states. Regarding the issue of hand, we are contrasting knowledge of our own beliefs gained through non-inferential, seemingly automatic means, on the one hand, and knowledge of our own beliefs gained through inferential or “inner observation”-based means, on the other hand. It’s a first-person issue through and through. So the difference between these two varieties of first-person self-knowledge can’t be explained by appeal to metaphysical differences; that isn’t guaranteed. Peacocke is right to look at epistemological considerations to address this issue.
Martin points out that while Peacocke claims that we form, often unconsciously, second-order beliefs just by looking out onto the world and gleaning first-order beliefs, he hasn’t explained exactly how this happens, or why directing our attention outwards results in such self-knowledge. In other words, while his proposal is plausible, that we have second-order beliefs just in having first-order beliefs, Peacocke hasn’t showed how this occurs.
Moreover, this proposal faces conceptual issues which Peacocke hasn’t considered. Martin points out that some of Peacocke’s reasoning dangerously relies upon phenomenological considerations. Self-knowledge of our beliefs can arise seemingly spontaneously as perceptual knowledge does. But phenomenology is under-determined regarding the nature of things. Namely, there is a certain distinction that Martin shows is central to the issues at hand, and it isn’t readily picked up via phenomenology, and rather relies upon conceptual consideration. There’s a distinction between something being an object of our attention, on the one hand, and something occupying our attention, on the other hand. For example, when the content of a thought is an object of our attention, we know that content; but when that content simply occupies our attention, all we can say is that we’re implicitly aware of having thought, but this need not imply that we could report on what the contents of that thought are. This distinction is particularly relevant when we talk about belief. In contrast, when we talk about perception, it does not matter as much because the notion of perception suggests the notion of relationality to the content of perception: when a perceptual experience occupies our attention, we are standing in relation to the perceptual object in the world, and so by necessity there’s an object of our attention. When thought and imagination enter the picture, we’re no longer in relation to their objects; there is no object of attention (in the relational sense) that having a thought or imagining guarantees, because they don’t have objects in the first place.
When it comes to Peacocke’s claim that we spontaneously inherit reasons, from looking at the world around us, which amount to the formation of a first-order belief and the guarantee that they can serve as reasons for second-order belief that we have that first-order belief, this distinction is critical. Looking at the world, as a perceptual affair, involves this relationality. But in forming beliefs about our beliefs, which can sometimes consist in primarily or only thought, we don’t have this relationality; it’s not like there’s a landscape of our beliefs set in stone and in a determinate form already there, and we respond to that. Martin thinks that Peacocke implicitly understands this and would hold that having thought consists in a determination of our attention, rather than there being thought as something independent of our attention and to which we direct our attention.
This distinction implies that more is required for a creature’s capacity to form second-order beliefs than the resources which suffice for forming perceptual knowledge. It shouldn’t be required that a creature possess the concept of beliefs; it’s natural to think that pre-linguistic humans can have awareness of what they believe. It also shouldn’t be that having a first-order belief itself constitutes a second-order belief regarding it; the two can come apart, for it is intuitive to think that sometimes knowledge of our beliefs require indirect routes, like reliance on the testimony of people who know us. But then what is the picture here? Martin points out that Peacocke doesn’t offer one. Instead, Peacocke only proposes, schematically, that the reasons for first-order belief are inherited as reasons for that belief to serve as a reason for a second-order belief in such a way that the essential conditions for the formation of that second-order belief are satisfied, and so it makes sense to say that we possess that second-order belief, in at least a standing, not occurrent, form.
Here’s another problem with Peacocke’s proposal. We don’t always look at the world in order to form first-order belief. Some of our first-order beliefs concern theoretical issues. Martin asks the reader to think about the example of thinking through a difficult philosophical problem. Here, we might not have a strong first-order belief about the matter (e.g., I don’t have a conclusive belief about the best way of understanding belief under psychological, rather than epistemological, contexts). Then, in order to know what we believe at the first-order, we might rely upon an inner feeling of conviction; we can propositionally imagine alternative positions and sense which one of them we feel most convinced by. In this case, we ought not expect that the reasons for holding that first-order belief transfer up to the level of the second-order belief; we don’t hold that second-order belief for the same reasons, for the new element of a feeling of conviction is part of the reason-base here.
In such cases, in what sense has the first-order belief served as a reason for holding the second-order belief? This is a problem for Peacocke’s view. Cases of self-deception, like the examples I gave above, can be understood in these terms as well. Such cases also bring out how it’s easy to get things less than right in theoretical cases. It’s very much possible that when we reflect upon a theoretical proposition in addressing a philosophical problem and we feel a strong conviction that, in fact, this conviction is responding to some other proposition which is related but distinct from this current one. In these cases, we can be less than trustworthy in our first-person beliefs about our beliefs, and this fact brings out just how tricky it is to demarcate the set of second-order beliefs, arrived at first-personally, which are trustworthy.
Martin points out that all we can say here is that there’s some “primitive rationally compelling transition” here, located between the first- and the second-order belief, and it should not be reduced to the perceptual kind. But given all of the problems which Peacocke’s view faces, Martin points out that given the current status of the debate, it’s not obvious or compelling that self-knowledge of our beliefs shouldn’t be reduced to that perceptual kind. Martin thinks Peacocke’s contribution to the debate is to make explicit that the proposal is that there’s such a primitive rationally compelling transition which is irreducible to the perceptual kind, and it leaves it to future thinkers to spell out what it consists in and to defend its existence.
Okie, so there’s a very rough overview of those two essays. Now, here are some of my ramblings in response to them. When I found myself first drawn to this topic of self-knowledge, I wasn’t quite sure why; on the surface, it doesn’t have much relation to what I’ve been focusing on (e.g., emotion regulation, and most locally, repressed emotions). Now, I see some potential connections. First, this debate shows that when we define what it means for some mental phenomenon to be conscious or unconscious, there are ways of doing it which have nothing to do with whether we are deliberately or occurrently transacting with it. Instead, it can be a matter of whether a “standing belief” exists about our having that mental phenomenon, where that mental state can be minimally understood in terms of that mental phenomenon properly figuring in as a personal reason for having that state such that we could report on it if we were appropriately prompted. It can also be a matter of how “inferentially integrated” that mental phenomenon is, which maybe can be understood as an independent consideration of that of whether we have a standing belief about it.
All of this is on top of Block’s famous distinction between access- and phenomenal-consciousness. What do Peacocke and Martin’s topic of study in these essays have to do with Block’s distinction? They are orthogonal; Peacocke’s account holds that in order to have a standing second-order belief (which could be glossed as access-conscious) one needs to be experiencing the world about (which could be glossed as phenomenally conscious) which one has the first-order belief which that second-order belief attributes to oneself. Other connections could be drawn, but they are independent matters. Block is focused on bringing out how relating to the mind phenomenally is quite different from relating to the mind via offering functional-level descriptions of its parts, whereas Peacocke and Martin in these essays are focused on addressing certain epistemological problems in such a way that it is compatible with their accounts to reject that those two types of description of the mind are basic, distinct, or exhaustive.
So to return to my main thought: I don’t think I should proceed in my writing by presupposing some sense of what it means for a mental phenomenon to be conscious or unconscious. That issue is just too complicated. Instead, I could proceed by talking about the trustworthiness of self-reports of the mental phenomena one is undergoing or possesses, where this trustworthiness could be understood in terms of either reliability or justifiability. In that case, when we speak of repressed emotions and desires, these could be understood in terms of that the subject has difficulty forming trustworthy self-reports about these, and we can thereby dodge the issue of whether the subject has “experienced” or “phenomenally encountered” the emotion or desire under consideration, or whether it is “accessible” in the sense that it can informally guide or control rational or behavioral activity or not.
Here’s another thought I have in response to the issues posed by Peacocke and Martin. I wonder whether the “primitive rationally compelling transition” which links first- and second-order beliefs must be open to variation in degree regarding how trustworthy and justifiable the resultant report on one’s belief is. Once the first-order belief under consideration is not a perceptual belief, and is rather partially constituted by imagination/thought qua a determination of one’s attention, it is possible that it is also stands as partially a determination of one’s interests and values. For example, when I think about philosophical issues, I am often interested in one thesis over another holding true, namely the one which would vindicate a position I’ve been married to for a while now. This is just a fancy way of saying that beliefs can be motivated or driven by desire. Maybe there’s one kind of primitive rationally compelling transition which defines cases in which our self-knowledge of our beliefs is transparent, automatic, or infallible, but the direction of this transition is highly reactive to desire/motivation/emotion. In some cases, this won’t be taken as impacting the validity of the knowledge because the starting first-order belief in the first place wasn’t totally formed, and so this “transition” is partially constitutive of the second-order belief; so questions of validity aren’t only essential, but it is also irreducibly partially a practical/ethical matter of what beliefs one self-constitutes. But in other cases, it will be taken as impacting the validity, namely those under which the starting rest-order belief was already well-formed.
I guess the authors might want to distinguish between knowledge of currently existent beliefs, on the one hand, and our making up our minds or changing our beliefs (to which coming to know our beliefs can sometimes amount to) on the other hand. My concern raised above confuses these two issues. One could rather say that when an extant belief is very general or non-committal, relative to beliefs at a more fine-grained level, which is the one that we're ultimately aiming to arrive at, it more effectively eludes our knowing it, or it will be especially responsible and malleable to our act of attempting to know it. So we shouldn't say that the primitive rationally compelling transition at hand admits of degrees in the way I suggested above; instead, when this kind of transition occurs, depending upon the level-of-grain of the starting belief involved, it can sometimes require that other sorts of cognitive activity, and even the subject's agency and desires, come into the picture, to help complete the transition. These musings indicate, however, that it's tricky; perhaps trying to specify the fine-grained processes involved in self-knowledge is futile, and instead we should stay at a rougher grain of analysis.
Reading these essays also provokes reflections on methodology in philosophy. Something seductive about the Peacocke-Martin exchange is that their approach gives the feeling of using a high-precision microscope, which focuses in on the subject of our knowledge of our beliefs. In contrast, other thinkers I’ve encountered on this subject matter can feel blunt, only skimming the surface (e.g., Krista Lawlor’s “Knowing what one wants”). I think this difference in feeling is based upon the sorts of considerations these different thinkers use in proceeding in argument. Lawlor’s paper consists in reflections upon a personal example. In contrast, Peacocke’s and Martin’s papers consist in raising questions of the commitments tied up to an intuitive claim, surveying the landscape of competing claims, and comparing their respective commitments. I think these different approaches each have an important role in the collective project of advancing our understanding of the issues at hand. Here’s a way of articulating the roles of these approaches. It’s important to look at phenomenology and personal examples for expanding the hypothesis space and for argumentation under the form of finding counterexamples of extant claims, but it’s critical to not draw conclusions between competing hypotheses without focusing on claims with respect to where they are nestled within implied conceptual networks.
Displaying 1 of 1 review