Saturday, September 6, 2008
Sunday, April 20, 2008
Phenomenalism I - Analytic vs. Ontological
Phenomenalism, as the name implies, has something to do with phenomena or appearances. What that something is turns out to vary to some degree depending on the theorist. Usually, when the moniker of phenomenalism is invoked it is to describe what I shall call analytic phenomenalism, which was a theory concerning the meaning of object statements and was held by empiricists from Mach to C.I. Lewis, Russell, and (early) Ayer. Analytic phenomenalism held that the meaning of ordinary object statements like “there is a desk in my study” were logically equivalent to statements about actual and possible sensations like “if I were in my study I would have deskish sensations”. These sensations were called sense data. What are sense data? They are the things we are (a) directly aware of in perception, (b) mind dependent, and (c) have the properties that they appear to have in our perception of them.
There are yet further varieties of phenomenalism. In contrast to analytic phenomenalism, ontological phenomenalism holds that objects just are collections of sense data. Berkeley and Mill held such a view. Berkeley puts it well when he says,
"As several [ideas] are observed to accompany each other, they come to be marked by one name, and so to be reputed as one thing. Thus, for example, a certain colour, taste, smell, figure, and consistence having been observed to go together, are accounted one distinct thing, signified by the name apple. Other collections of ideas constitute a stone, a tree, a book, and the like sensible things."
Ontological phenomenalism clearly remains within the bounds of Cartesian philosophical inquiry. Unlike analytical phenomenalism it evades skeptical problems about the inferred existence of an external world by simply closing the gap. There is no world of objects distinct from the sensory contents of consciousness whose existence must be inferred or posited. Therefore there is no ‘veil of perception’ and no skeptical problem. Hence, ontological phenomenalism adds an additional thesis to the constellation of doctrines held by analytical phenomenalism. This is (d) sense data have no representational or intentional properties. In other words, sense-data do not point beyond themselves to anything else. They are non-representational, epistemically (and in Mill’s case, metaphysically) fundamental entities. It also makes, (b) mind dependence, an optional part of the doctrine.
A few further things need to be said about phenomenalism generally. First, phenomenalism is not committed to an ontology of sense data. Sensation can be explained in other ways. Once can just as easily be a phenomenalist while holding, for instance, that sensations are adverbial modifications of states of perceivers. Second, the analytic and ontological positions are logically independent of one another. One might hold an ontological phenomenalism and yet think that all object statements are about non-phenomenal objects and hence in error. This would be akin to a Mackien error theory for statements concerning the objects of perception. Likewise, analytic phenomenalism can be held while thinking that objects ultimately are some theoretically posited, super-sensible entities (e.g. beyond the ‘veil of perception’), even though prosaic statements about perceptual objects are not about such entities.
There are yet further varieties of phenomenalism. In contrast to analytic phenomenalism, ontological phenomenalism holds that objects just are collections of sense data. Berkeley and Mill held such a view. Berkeley puts it well when he says,
"As several [ideas] are observed to accompany each other, they come to be marked by one name, and so to be reputed as one thing. Thus, for example, a certain colour, taste, smell, figure, and consistence having been observed to go together, are accounted one distinct thing, signified by the name apple. Other collections of ideas constitute a stone, a tree, a book, and the like sensible things."
Ontological phenomenalism clearly remains within the bounds of Cartesian philosophical inquiry. Unlike analytical phenomenalism it evades skeptical problems about the inferred existence of an external world by simply closing the gap. There is no world of objects distinct from the sensory contents of consciousness whose existence must be inferred or posited. Therefore there is no ‘veil of perception’ and no skeptical problem. Hence, ontological phenomenalism adds an additional thesis to the constellation of doctrines held by analytical phenomenalism. This is (d) sense data have no representational or intentional properties. In other words, sense-data do not point beyond themselves to anything else. They are non-representational, epistemically (and in Mill’s case, metaphysically) fundamental entities. It also makes, (b) mind dependence, an optional part of the doctrine.
A few further things need to be said about phenomenalism generally. First, phenomenalism is not committed to an ontology of sense data. Sensation can be explained in other ways. Once can just as easily be a phenomenalist while holding, for instance, that sensations are adverbial modifications of states of perceivers. Second, the analytic and ontological positions are logically independent of one another. One might hold an ontological phenomenalism and yet think that all object statements are about non-phenomenal objects and hence in error. This would be akin to a Mackien error theory for statements concerning the objects of perception. Likewise, analytic phenomenalism can be held while thinking that objects ultimately are some theoretically posited, super-sensible entities (e.g. beyond the ‘veil of perception’), even though prosaic statements about perceptual objects are not about such entities.
ENAKS conference
Just got back from a weekend in NYC. Had a great time at the Eastern division meeting of the North American Kant Society. Though my brain feels like tapioca I got to hear a bunch of interesting papers on Kant and meet a lot of great people. I'll post a list of the papers that were given when I get a chance.
There's also a few posts on phenomenalism and whether Kant and Leibniz were phenomenalists that i'm going to try to put up soon
There's also a few posts on phenomenalism and whether Kant and Leibniz were phenomenalists that i'm going to try to put up soon
Thursday, July 19, 2007
Conceptualism and Blind Reasoning
Here's an argument I've been working on, based in part on Bill Brewer's essay "Compulsion by Reason", for thinking that only conceptually structured representational content can play a role in rational thought:
Reasons, in the appropriate sense, have an articulation requirement such that the agent be able not merely to act in accord with a given reason but rather act on account of that very reason. The idea is that whenever certain transitions, whether in thought or action, are made in a rational manner, such transitions accord with certain relevant norms prescribing how things are to be done. Given this general idea, there is always a distinction to be made between the subject’s being in mere accordance with the norm and his being guided by an understanding of the norm. It is only in the latter situation that it is appropriate to ascribe to the subject possession of a reason for making the transition.
Hence, possession of a reason requires a certain conscious awareness of what one is doing in making a transition and why it is that he is transitioning in this way rather than that. Denying this claim makes it difficult to see how the reason could play a rational as opposed to merely causal role in the fixation of belief. Put another way, we might say that in order for something to count as being a reason capable of being acknowledged by the agent it must present itself in a certain manner such that its significance is capable of being grasped by the thinker and put to use in a process of inferential or probabilistic reasoning. We can illustrate this by considering the problem of blind reasoning.
What is involved in following a valid deductive argument with real understanding, in a way that yields knowledge of its conclusion? Putting aside the possibility of a slip or unnoticed error, in following a valid deductive argument and understanding it as valid one is correctly compelled by the argument to believe its conclusion. Moreover, in following and fully understanding the argument, this compulsion is not simply a blind and mysterious manipulation of one’s beliefs by some reliable mechanism, however well established by evolution, benevolent brain washing, etc. If one’s understanding of the argument is really to extend one’s knowledge, then that understanding must give one some appreciation of why one is right in believing its conclusion. One has to have some grasp of how the conclusion follows from the premises, i.e. of how one knows the conclusion, and one’s resultant belief should be guided by this understanding. As Bill Brewer says (in "Compulsion by Reason"),
"A disposition to take beliefs on board in parallel with the steps of the argument is on its own insufficient for the argument to provide me with genuine knowledge. For such beliefs would come as a succession of mere hunches, wholly unsubstantiated for me by the de facto validity of the argument propelling my endorsement of them."
There is thus more to grasping the laws of logic or mathematical argument or even to understanding a language than simply being disposed to have one’s beliefs mirror the moves which those laws prescribe. Hence, reasoning that demonstrates genuine understanding, reasoning that is demonstrably done for reasons, is not a merely mechanical manipulation of belief, but, to use Brewer’s phrase, “a compulsion in thought by reason”, and as such involves some conscious understanding of why one is right in one’s conclusions. The knowledge which drives one in cases of genuine understanding is precisely what one lacks if one has simply been drilled by the dictators of the International Academy of Logic, however benevolent they may be, to reason in ways for which one sees no evident rationale, no point or purpose.
Avoiding the problem of blind reasoning cannot be accomplished by adverting to explicit rules which an agent consciously follows in making a transition. The agent must have some coherent grasp of the point and purpose of the transitions he is making, over and above mere inclination or compulsion, which motivates making the correct transition on particular occasions and makes sense for the agent of why he is right in doing as he does. Nevertheless, this cannot consist in any explicit knowledge of an abstract general algorithm for making a correct transition, or full-blown reflective understanding of the norms governing the practice in question, independent of the agent’s actual behavior in conforming to it. The problem, then, is how to characterize this crucial mode of awareness.
Generally such characterizations of the norms governing cognitive capacities oscillate between two extremes: (1) the Platonic conception of infinitely extended rails which define correctness on every possible occasion completely independently of a competent rule-follower’s inclination to go on, and to which the rule-follower is attached by explicit knowledge of their course at every point; and (2) the idea that understanding of a rule might be reduced to an actual user’s dispositions to make a particular transition when confronted with some environmental stimulus. The discussion of blind reasoning is meant to undermine both of these conceptions. What is required is an account of the causal dependence of rational transitions on reasons which itself makes essential appeal to some grasp on the part of the subject of the point of what he is doing and why he is right in doing it. This account would give an integrated explanation of what it is to grasp a rule or norm that would not require positing a third realm or platonic heaven whose denizens are grasped via some spooky perceptual analogue. This intrinsic and motivating grasp of the reasonableness of his behavior, which makes the causal dependence one in virtue of rationalization by reasons, should nevertheless be quite different from any independent reflective understanding or explicit knowledge of a general rule governing the practice in question, from which the appropriate action is to be inferred on any particular occasion. Exactly how causal dependence in virtue of rationalization along these lines should be elucidated in each case in which it is operative is a difficult and important matter.
The conceptualist constraint upon rationalization by reasons is then the following: Whatever one’s positive account of a rational transition, such an account must provide some basis for the intelligibility of the transition from the subject’s point of view. Hence, given this articulation of what it is for a consideration to be a reason for a particular transition, i.e. that it be capable of playing a role in inference, and recognizable as such by the thinker, reasons must be conceptual. Therefore, if representational content is to provide reasons that content must be conceptual in nature.
Reasons, in the appropriate sense, have an articulation requirement such that the agent be able not merely to act in accord with a given reason but rather act on account of that very reason. The idea is that whenever certain transitions, whether in thought or action, are made in a rational manner, such transitions accord with certain relevant norms prescribing how things are to be done. Given this general idea, there is always a distinction to be made between the subject’s being in mere accordance with the norm and his being guided by an understanding of the norm. It is only in the latter situation that it is appropriate to ascribe to the subject possession of a reason for making the transition.
Hence, possession of a reason requires a certain conscious awareness of what one is doing in making a transition and why it is that he is transitioning in this way rather than that. Denying this claim makes it difficult to see how the reason could play a rational as opposed to merely causal role in the fixation of belief. Put another way, we might say that in order for something to count as being a reason capable of being acknowledged by the agent it must present itself in a certain manner such that its significance is capable of being grasped by the thinker and put to use in a process of inferential or probabilistic reasoning. We can illustrate this by considering the problem of blind reasoning.
What is involved in following a valid deductive argument with real understanding, in a way that yields knowledge of its conclusion? Putting aside the possibility of a slip or unnoticed error, in following a valid deductive argument and understanding it as valid one is correctly compelled by the argument to believe its conclusion. Moreover, in following and fully understanding the argument, this compulsion is not simply a blind and mysterious manipulation of one’s beliefs by some reliable mechanism, however well established by evolution, benevolent brain washing, etc. If one’s understanding of the argument is really to extend one’s knowledge, then that understanding must give one some appreciation of why one is right in believing its conclusion. One has to have some grasp of how the conclusion follows from the premises, i.e. of how one knows the conclusion, and one’s resultant belief should be guided by this understanding. As Bill Brewer says (in "Compulsion by Reason"),
"A disposition to take beliefs on board in parallel with the steps of the argument is on its own insufficient for the argument to provide me with genuine knowledge. For such beliefs would come as a succession of mere hunches, wholly unsubstantiated for me by the de facto validity of the argument propelling my endorsement of them."
There is thus more to grasping the laws of logic or mathematical argument or even to understanding a language than simply being disposed to have one’s beliefs mirror the moves which those laws prescribe. Hence, reasoning that demonstrates genuine understanding, reasoning that is demonstrably done for reasons, is not a merely mechanical manipulation of belief, but, to use Brewer’s phrase, “a compulsion in thought by reason”, and as such involves some conscious understanding of why one is right in one’s conclusions. The knowledge which drives one in cases of genuine understanding is precisely what one lacks if one has simply been drilled by the dictators of the International Academy of Logic, however benevolent they may be, to reason in ways for which one sees no evident rationale, no point or purpose.
Avoiding the problem of blind reasoning cannot be accomplished by adverting to explicit rules which an agent consciously follows in making a transition. The agent must have some coherent grasp of the point and purpose of the transitions he is making, over and above mere inclination or compulsion, which motivates making the correct transition on particular occasions and makes sense for the agent of why he is right in doing as he does. Nevertheless, this cannot consist in any explicit knowledge of an abstract general algorithm for making a correct transition, or full-blown reflective understanding of the norms governing the practice in question, independent of the agent’s actual behavior in conforming to it. The problem, then, is how to characterize this crucial mode of awareness.
Generally such characterizations of the norms governing cognitive capacities oscillate between two extremes: (1) the Platonic conception of infinitely extended rails which define correctness on every possible occasion completely independently of a competent rule-follower’s inclination to go on, and to which the rule-follower is attached by explicit knowledge of their course at every point; and (2) the idea that understanding of a rule might be reduced to an actual user’s dispositions to make a particular transition when confronted with some environmental stimulus. The discussion of blind reasoning is meant to undermine both of these conceptions. What is required is an account of the causal dependence of rational transitions on reasons which itself makes essential appeal to some grasp on the part of the subject of the point of what he is doing and why he is right in doing it. This account would give an integrated explanation of what it is to grasp a rule or norm that would not require positing a third realm or platonic heaven whose denizens are grasped via some spooky perceptual analogue. This intrinsic and motivating grasp of the reasonableness of his behavior, which makes the causal dependence one in virtue of rationalization by reasons, should nevertheless be quite different from any independent reflective understanding or explicit knowledge of a general rule governing the practice in question, from which the appropriate action is to be inferred on any particular occasion. Exactly how causal dependence in virtue of rationalization along these lines should be elucidated in each case in which it is operative is a difficult and important matter.
The conceptualist constraint upon rationalization by reasons is then the following: Whatever one’s positive account of a rational transition, such an account must provide some basis for the intelligibility of the transition from the subject’s point of view. Hence, given this articulation of what it is for a consideration to be a reason for a particular transition, i.e. that it be capable of playing a role in inference, and recognizable as such by the thinker, reasons must be conceptual. Therefore, if representational content is to provide reasons that content must be conceptual in nature.
Tuesday, July 17, 2007
John McDowell on Reasons and Knowledge
Here's a part of a paper I've been working on - this particular argument concerns a problem I see with McDowell's conception of justification and knowledge:
It may be easiest to describe McDowell’s position by discussing what it is not, namely a simplistic hybridization of an 'internalist' conception of justification with certain 'externalist' flourishes. This hybrid conception combines an internally or subjectively accessible notion of a reason with an externalist conception of entitlement, where such entitlement is based on the truth preserving nature of the transitions it entitles, but defines them each in such a way that they are entirely distinct and independent from one another. The internalist notion of a reason is then something the subject can ensure is met. It is solely a matter of showing how the consideration can play a role in the subject’s reasoning. Acknowledgment of such reasons is a necessary condition of the subject’s judgment providing knowledge of its content. The entitlement, on the other hand, is something additional to reasons, and is blankly external to the subject. The individual need not have the concepts necessary to think the propositional content that formulates the warrant. It shows the way in which the world, and the subject’s place in the world, has to be if his judgments are to provide knowledge. It shows the conditions under which the subject’s judgments are true.
McDowell sees two difficulties with this account. The first is that by separating reasons from truth it makes it difficult to see how citing something as reason can satisfy a role it is surely has in reasoning, namely that one’s reasons are true. Hence the consideration of what reasons one has cannot be something entirely internal to how things are with the subject. If it were it would be difficult to see how reasons could be connected to truth in the way which they intuitively seem to be.
Second, if entitlement and reason are distinct in the way outlined it seems to make their connection an entirely accidental circumstance, one where the obtaining of conditions necessary for entitlement are themselves completely independent of the internally constituted reasons which the subject cites in making his judgments. Thus whether the external conditions do obtain is a matter of indifference from the point of view of the subject. This sort of accidental relation between reasons and entitlement seems insufficient to ground knowledge.
McDowell’s solution is to deny that any hybridization of entitlement and reason is possible. Instead, he reconceives a reason as synonymous with a fact. What reasons a subject has then depends upon how things are with the world. He thus rejects the idea that “reason must be credited with a province within which it has absolute control over the acceptability of positions achievable by its exercise, without laying itself open to risk from an unkind world”. Or as he puts it elsewhere, what reasons a subject has depends upon the world’s "doing the subject a favor". In other words, there are no considerations which can count as reasons for an agent which are not true. To think otherwise is to accept a natural but mistaken view of epistemic standing, one credited by McDowell with deeply Cartesian roots.
The truth of McDowell’s account cannot be assessed here. What is of immediate interest for our purposes is that McDowell’s account of knowledge seems to privilege entitlement over possession of a reflectively accessible reason in describing the epistemic standing of the subject. However there are two aspects in which McDowell’s position might be construed as distinct from a 'pure' entitlement theory. First, McDowell is concerned with showing how reasons can be dependent upon how the world is, i.e., with truth, without their being “blankly external” to the subject who has them. Second, what reasons a subject has still depends upon the subject’s conceptual capacities. Hence, even if the accessibility to reasons is not equivalent to reflective accessibility, the subject’s epistemic standing cannot be assessed independently of the conceptual capacities of the subject. We’ll examine these two aspects in turn.
We can understand McDowell’s worry about the potential blank externality of reasons to one’s epistemic standing by first glossing what he sees as a natural way of thinking about justification.
"The root idea [of the gripping internalism] is that one’s epistemic standing on some question cannot intelligibly be constituted, even in part, by matters blankly external to how it is with one subjectively. For how could such matter be other than beyond one’s ken? And how could matters beyond one’s ken make any difference to one’s epistemic standing?"
To understand the significance of this passage we must engage in some, always difficult, McDowell exegesis. The three problematic notions in this passage are ‘epistemic standing’, ‘blankly external’ and ‘beyond one’s ken’. ‘Epistemic standing’ I take as describing when one is in a position to know that some fact p obtains. For a fact to be ‘beyond one’s ken’ is for one not to be in a position to know that p obtains. Finally, for a fact p to be ‘blankly external’ to one is for it to be irrelevant to one’s subjective states whether or not p is true.
On this interpretation we can gloss the passage as saying that (1) Blankly external matters are beyond one’s ken; (2) Matters beyond one’s ken make no difference to one’s epistemic standing; therefore (3) Blankly external matters cannot make any difference to one’s epistemic standing. This clearly sounds like a very strong form of internalism and seems to capture adequately the internalist's idea that only considerations ‘within one’s ken’ make any difference to what reasons one has.
McDowell’s ingenious move here is to maintain (1-3) while denying that how things are with the world must be blankly external to one, and thereby beyond one’s ken. He does this by arguing that, at least sometimes, one knows that p because p is made "manifest" to one.
"When someone has a fact made manifest to him, the obtaining of the fact contributes to his epistemic standing on the question. But the obtaining of the fact is precisely not blankly external to his subjectivity, as it would be if the truth [about his subjectivity] were exhausted by the highest common factor [of his experience that p, namely its appearing to him that p]."
What is it for a fact to be "made manifest" to one? It is for a subject to be in a mental state, phenomenologically indistinguishable from its merely appearing to him that p, which entails that p is the case. What makes the mental state factive in this way is that it is individuated externally, that is, the conditions that determine which M one is using supervene, at least partly, on the environment external to the agent at some particular time. Since part of what it is for one to be in mental state is to stand in certain relations to one’s environment it can both be the case that a fact is not blankly external to one and that that fact be (at least for the moment) reflectively inaccessible to one. Hence, when a fact is made manifest to one, one is in good epistemic standing – one knows that p. What started out (with the ‘gripping internalism’) as a very internalist picture of knowledge can now be shown compatible with a very externalist conception of the mind. Indeed, McDowell’s ingenious move is to make content externalism work in the service of epistemological internalism.
If the above account is correct we are now in a position to see how McDowell’s conception of knowledge is connected with the primary concern of this paper, the dispute between conceptualists and non-conceptualists about reasons and rationality. What needs explanation is how McDowell can advance a conceptualist thesis that seems to hinge around the accessibility and articulability of a subject’s reasons while at the same time holding to a conception of knowledge as something that does not require such subjective access, though it does require that the subject’s knowledge not be blankly external to him.
This tension is expressible in the form of a dilemma: either accept an account of knowledge that allows one’s reasons need not be reflectively accessible, thereby giving up the most significant argument against non-conceptual content, or admit that reasons must be reflectively accessible thereby giving up the account of knowledge as facts ‘made manifest’ to thinker. It seems unlikely that McDowell would be willing to modify his account of knowledge, so the question is how best to characterize his conceptualist position such that it is tenable as a coherent challenge to non-conceptualism.
It is not at all clear that any intermediary position is available to McDowell. His theory of knowledge requires giving up the notion that one’s reasons must, qua reasons, be able to play a role in one’s deliberations about what to believe. Since the manifest nature of a fact that p is indistinguishable, from the subject’s point of view, from its merely appearing that p, there is no way to ensure that what the subject takes to be reasons, when reflecting upon what to believe, really are reasons. This makes it altogether uncertain how it could ever be the case that a subject’s reasons, taken as consisting of something more that how things seem to the subject, could ever play the active role in thinking that was discussed in section II.
A partial response on behalf of McDowell is to emphasize that on his account, one’s reasons are still epistemically accessible to the subject – after all they are not blankly external to him – but they are just not reflectively accessible. But when p is made manifest to a thinker and hence known, that state of knowledge is itself a mental one and thereby available to play a rational role in the fixation of belief. Hence, so long as one has epistemic access to one’s reasons they will be available to play a role in the fixation of belief even if they are not reflectively understood as having that role.
This is only a partial response because though it shows how reasons can still be causes it does not show how one’s reasons can act as guides in making correct transitions in thought. Indeed, McDowell’s externalism seems to run into the very same charge of blind reasoning that was brought against the non-conceptualist. The strongest argument that conceptualism possesses for its own position and against that of the non-conceptualist seems to hinge on a thoroughly internalist conception of a reason. McDowell’s disjunctivist conception of knowledge makes it altogether unclear how a subject is able to grasp her reasons, qua reasons, in such a way that they can guide her beliefs or actions. It seems unlikely that McDowell’s theory, as it now stands, has the resources to avoid this problem.
This problem for McDowell makes me think that the conceptualist position must hold something like the hybrid conception of reasons and entitlement but try to do so in such a way that one’s reasons need not be, in principle, independent of the truth. Admittedly, it is difficult to see how exactly this would work. Part of the task of successfully articulating a plausible conceptualist position requires solving this problem.
It may be easiest to describe McDowell’s position by discussing what it is not, namely a simplistic hybridization of an 'internalist' conception of justification with certain 'externalist' flourishes. This hybrid conception combines an internally or subjectively accessible notion of a reason with an externalist conception of entitlement, where such entitlement is based on the truth preserving nature of the transitions it entitles, but defines them each in such a way that they are entirely distinct and independent from one another. The internalist notion of a reason is then something the subject can ensure is met. It is solely a matter of showing how the consideration can play a role in the subject’s reasoning. Acknowledgment of such reasons is a necessary condition of the subject’s judgment providing knowledge of its content. The entitlement, on the other hand, is something additional to reasons, and is blankly external to the subject. The individual need not have the concepts necessary to think the propositional content that formulates the warrant. It shows the way in which the world, and the subject’s place in the world, has to be if his judgments are to provide knowledge. It shows the conditions under which the subject’s judgments are true.
McDowell sees two difficulties with this account. The first is that by separating reasons from truth it makes it difficult to see how citing something as reason can satisfy a role it is surely has in reasoning, namely that one’s reasons are true. Hence the consideration of what reasons one has cannot be something entirely internal to how things are with the subject. If it were it would be difficult to see how reasons could be connected to truth in the way which they intuitively seem to be.
Second, if entitlement and reason are distinct in the way outlined it seems to make their connection an entirely accidental circumstance, one where the obtaining of conditions necessary for entitlement are themselves completely independent of the internally constituted reasons which the subject cites in making his judgments. Thus whether the external conditions do obtain is a matter of indifference from the point of view of the subject. This sort of accidental relation between reasons and entitlement seems insufficient to ground knowledge.
McDowell’s solution is to deny that any hybridization of entitlement and reason is possible. Instead, he reconceives a reason as synonymous with a fact. What reasons a subject has then depends upon how things are with the world. He thus rejects the idea that “reason must be credited with a province within which it has absolute control over the acceptability of positions achievable by its exercise, without laying itself open to risk from an unkind world”. Or as he puts it elsewhere, what reasons a subject has depends upon the world’s "doing the subject a favor". In other words, there are no considerations which can count as reasons for an agent which are not true. To think otherwise is to accept a natural but mistaken view of epistemic standing, one credited by McDowell with deeply Cartesian roots.
The truth of McDowell’s account cannot be assessed here. What is of immediate interest for our purposes is that McDowell’s account of knowledge seems to privilege entitlement over possession of a reflectively accessible reason in describing the epistemic standing of the subject. However there are two aspects in which McDowell’s position might be construed as distinct from a 'pure' entitlement theory. First, McDowell is concerned with showing how reasons can be dependent upon how the world is, i.e., with truth, without their being “blankly external” to the subject who has them. Second, what reasons a subject has still depends upon the subject’s conceptual capacities. Hence, even if the accessibility to reasons is not equivalent to reflective accessibility, the subject’s epistemic standing cannot be assessed independently of the conceptual capacities of the subject. We’ll examine these two aspects in turn.
We can understand McDowell’s worry about the potential blank externality of reasons to one’s epistemic standing by first glossing what he sees as a natural way of thinking about justification.
"The root idea [of the gripping internalism] is that one’s epistemic standing on some question cannot intelligibly be constituted, even in part, by matters blankly external to how it is with one subjectively. For how could such matter be other than beyond one’s ken? And how could matters beyond one’s ken make any difference to one’s epistemic standing?"
To understand the significance of this passage we must engage in some, always difficult, McDowell exegesis. The three problematic notions in this passage are ‘epistemic standing’, ‘blankly external’ and ‘beyond one’s ken’. ‘Epistemic standing’ I take as describing when one is in a position to know that some fact p obtains. For a fact to be ‘beyond one’s ken’ is for one not to be in a position to know that p obtains. Finally, for a fact p to be ‘blankly external’ to one is for it to be irrelevant to one’s subjective states whether or not p is true.
On this interpretation we can gloss the passage as saying that (1) Blankly external matters are beyond one’s ken; (2) Matters beyond one’s ken make no difference to one’s epistemic standing; therefore (3) Blankly external matters cannot make any difference to one’s epistemic standing. This clearly sounds like a very strong form of internalism and seems to capture adequately the internalist's idea that only considerations ‘within one’s ken’ make any difference to what reasons one has.
McDowell’s ingenious move here is to maintain (1-3) while denying that how things are with the world must be blankly external to one, and thereby beyond one’s ken. He does this by arguing that, at least sometimes, one knows that p because p is made "manifest" to one.
"When someone has a fact made manifest to him, the obtaining of the fact contributes to his epistemic standing on the question. But the obtaining of the fact is precisely not blankly external to his subjectivity, as it would be if the truth [about his subjectivity] were exhausted by the highest common factor [of his experience that p, namely its appearing to him that p]."
What is it for a fact to be "made manifest" to one? It is for a subject to be in a mental state, phenomenologically indistinguishable from its merely appearing to him that p, which entails that p is the case. What makes the mental state factive in this way is that it is individuated externally, that is, the conditions that determine which M one is using supervene, at least partly, on the environment external to the agent at some particular time. Since part of what it is for one to be in mental state is to stand in certain relations to one’s environment it can both be the case that a fact is not blankly external to one and that that fact be (at least for the moment) reflectively inaccessible to one. Hence, when a fact is made manifest to one, one is in good epistemic standing – one knows that p. What started out (with the ‘gripping internalism’) as a very internalist picture of knowledge can now be shown compatible with a very externalist conception of the mind. Indeed, McDowell’s ingenious move is to make content externalism work in the service of epistemological internalism.
If the above account is correct we are now in a position to see how McDowell’s conception of knowledge is connected with the primary concern of this paper, the dispute between conceptualists and non-conceptualists about reasons and rationality. What needs explanation is how McDowell can advance a conceptualist thesis that seems to hinge around the accessibility and articulability of a subject’s reasons while at the same time holding to a conception of knowledge as something that does not require such subjective access, though it does require that the subject’s knowledge not be blankly external to him.
This tension is expressible in the form of a dilemma: either accept an account of knowledge that allows one’s reasons need not be reflectively accessible, thereby giving up the most significant argument against non-conceptual content, or admit that reasons must be reflectively accessible thereby giving up the account of knowledge as facts ‘made manifest’ to thinker. It seems unlikely that McDowell would be willing to modify his account of knowledge, so the question is how best to characterize his conceptualist position such that it is tenable as a coherent challenge to non-conceptualism.
It is not at all clear that any intermediary position is available to McDowell. His theory of knowledge requires giving up the notion that one’s reasons must, qua reasons, be able to play a role in one’s deliberations about what to believe. Since the manifest nature of a fact that p is indistinguishable, from the subject’s point of view, from its merely appearing that p, there is no way to ensure that what the subject takes to be reasons, when reflecting upon what to believe, really are reasons. This makes it altogether uncertain how it could ever be the case that a subject’s reasons, taken as consisting of something more that how things seem to the subject, could ever play the active role in thinking that was discussed in section II.
A partial response on behalf of McDowell is to emphasize that on his account, one’s reasons are still epistemically accessible to the subject – after all they are not blankly external to him – but they are just not reflectively accessible. But when p is made manifest to a thinker and hence known, that state of knowledge is itself a mental one and thereby available to play a rational role in the fixation of belief. Hence, so long as one has epistemic access to one’s reasons they will be available to play a role in the fixation of belief even if they are not reflectively understood as having that role.
This is only a partial response because though it shows how reasons can still be causes it does not show how one’s reasons can act as guides in making correct transitions in thought. Indeed, McDowell’s externalism seems to run into the very same charge of blind reasoning that was brought against the non-conceptualist. The strongest argument that conceptualism possesses for its own position and against that of the non-conceptualist seems to hinge on a thoroughly internalist conception of a reason. McDowell’s disjunctivist conception of knowledge makes it altogether unclear how a subject is able to grasp her reasons, qua reasons, in such a way that they can guide her beliefs or actions. It seems unlikely that McDowell’s theory, as it now stands, has the resources to avoid this problem.
This problem for McDowell makes me think that the conceptualist position must hold something like the hybrid conception of reasons and entitlement but try to do so in such a way that one’s reasons need not be, in principle, independent of the truth. Admittedly, it is difficult to see how exactly this would work. Part of the task of successfully articulating a plausible conceptualist position requires solving this problem.
the first
post that is...this blog is still very much under construction but i'll be posting (at some point) on whatever i feel like until someone says otherwise. It will probably be some academic stuff - reader comments will be welcome.
Subscribe to:
Posts (Atom)
