Course: Contemporary European Philosophy: Perception Professor: Frank Hofmann Master Course W/S 2014 Université du Luxembourg
McDowell's "Perception as a Capacity for Knowledge"
Name: Scheer Kris Email:
[email protected] Student_ID: 0070394500
The aim of this paper is to present McDowell's paper "Perception as a Capacity for Knowledge" which was published in 2011. I will present the main line of argument aswell as discuss some major problems that result from it. I will try to give a clear and structured account of the theses and antitheses discussed in order to ultimately focus on the falibility problem of perception. The falibility of a perceptual state and the resulting problems for perceptual knowledge and by extend perceptual justification is the focal point of McDowell's paper and it will therefore be the center of attention for this paper aswell. I will however start with explaining the conception of knowledge as it is understood and used by McDowell and contrast it with Burge's, one of the critics cited by McDowell himself. From there the attention will shift towards Burge's objection concerning the falibility of perception. In his paper „Perception as Capacity for Knwoledge“, McDowell argues for a thesis according to which perception can represent a viable warrant for the believes of a subject about the states of the world. He does so by way of criticism of Burge and by leaning on the theories of Sellars. First it is necessary to outline McDowell’s and by extension Sellar’s position regarding the discussion: McDowell is primary concerned with a very distinctive kind of knowledge which is orientated around what Sellars calls „self-concious rationality at work“. This kind of observational knowledge includes a certain selfconciousness about the credentials of beliefs. Additionally for Sellars, rationality is always closely related to language. In other terms: to be a rational animal is the same as being a language-using animal. Thus acts of reason can only be performed by language using animals in addition to them being selfconscious about the credentials of their knowledge. If a person has some kind of knowledge, he can not only state what he knowledgeably believes but he can aditionally state how that belief is rationally grounded. This specific kind of knowledge as rationality in operation, or acts of reason, does not include non-human animals or pre-linguistic children. This Sellarsian conception of knowldge, where the warrant through which a belief counts as knowledge is accessible to, i.e. potentially known by the subject, is called epistemological „internalism“. In this sense someone who has a bit of knwoledge is self-conscious about the credentials of his knowledge; his warrants are accessible to him. It is at this pivotal point of Sellars’ theory that Tyler Burges, an epistemological externalist, focuses his first attack. He does so by differenciating between what he calls „entitlement“ and „justification“ as two kinds of warrants. The former indicates a kind of warrant which „needs not be fully conceptually accessible, even on
reflection, to the warranted individual.“1 Justification on the other hand indicates a warrant that is conceptually accessible. In detail2, Burges divides warrants into two sub-species: Justification which is epistemically internalist due to its being warranted by reason that is conceptually accessible upon reflection. Beliefs that are justified thus result from reasoning or from inference from other beliefs. Burges now holds that when it comes to perceptual belief, there is no inference from the perceptual state to an actual belief. In other words in most typical situations we do not reason our way from experience to belief. Consequently the warrant for a perceptual beleif needs to be of a different kind, namely: entitlement. Whereas justification was epistemologically internalist, entitlement is externalist in that it need not be conceptually accessible. Entitlements are beliefs that are unsupported by evidence available to the subject but which the subject has still the right to hold epistemically. A subject being entitled is determined by facts and circumstances which the subject need not understand or be able to recognize. For Burges perceptual states only provide this first kind of warrant for beliefs about the state of the world. Burges main motivation for this is the idea that perceptual knowledge can be had by non-human animals and prelinguistic humans. He urges for a theory that gives an account of perceptual warrants that suits all cases of perceptual knowledge. However, McDowell argues, that ascribing a special kind of perceptual knowledge to rational animals is not inconsistent with the idea that perceptual knowledge can be instantiated more primitively in non-rational animals or prelinguistic human children. Burges however goes on denying that the standards demanded by Sellarsian theory can be upheld even by most rational human beings. He does acknowledge that sometimes rational beings can „justify“3 their perceptual beliefs although most of the time the only warrant that one has is „entitlement“. He simply denies that ordinary people have the kind of conceptual sophistication that is necessary for „justification“. He further argues that although there may be people who are conceptually sophisticated enough as to make their warrant conceptually accessible to them, that does not guarantee that the belief itself is indefeasible. If one was to justify a perceptual belief by refering to a selfconsciously possessed warrant one would for exemple claim: „I saw such and such“. The warrant for this claim is the perceptual state that is selfconsciously accessible to the person uttering the statement. But how is this capable of warranting the perceptual belief? Lets consider a non-veridical perceptual state like an illusion: the epistemic contribution offered by 1
McDowell, John.: Perception as a capacity for knowledge. S.19 http://www.iep.utm.edu/ep-en/#H2 3 here « justification » is used in Burges terms as meaning a warrant that is conceptually accessible on reflection to the warranted individual. 2
such an non-veridical perceptual state is none. Thus Burges main line of attack is the claim that perception cannot indefeasibly warrant our beliefs about the world because perception can be deceived and therefore perception can only provide us with an inconclusive warrant for our beliefs. McDowell however argues, that if all goes well in the operation of a perceptual capacity, a perceiver enjoys a perceptual state in which feautures of his environment are actually there for him. The feautures are perceptually present to her rationally self-conscious awareness.4 And if this is the case, that a perceiver has a feauture of the environment perceptually present to him, a perceptual state can guarantee the truth of a believe. According to McDowell, "if a perceptual state makes a feauture of the environment present to a perceivers's rationally self-conscious awareness, there is no possibility, compatible with someone's being in that state, that things are not as the state would warrant him in believing that they are."5 A case such as Burges describes, a case of non-veridical perception, is not what McDowell describes as being "experiences of". It does not follow from the fact that perceptions can be fallible, that perception, in the good case, cannot be a true warrant for a belief. Burges is right in saying that having a perceptional warrant is compatible with being wrong in the warranted belief (being fooled) but according to McDowell he is wrong in ascerting that a perceptual warrant is in itself neutral as to wether a belief it warrants is true or not. McDowell holds, that perception which is already conceptually shaped has an intinsic truth value, eg. in normal cases the ground that perception offers for a belief is a veridical one. This misunderstanding however seems to McDowell to reflect a mistake about the concept of fallibility, a mistake that he thinks is pervasive in epistemology in general. Fallibility is a property of capacities. It does not however put any restrictions on what a fallible capacity can be. Of course our perceptual capacity is fallible but nontheless it is a capacity to get into states that consist in having a certain feature of the world perceptually present to one's self conscious awareness.6 That is what this capacity does in a non-defective exercise of it. In such a case, it is impossible that things are not as the observer would believe them to be. McDowell puts it this way: "When we acknowledge that a capacity is fallible, we acknowledge that there can be exercises of it that are defective, in that they fail to be cases of what the capacity is specified as a capacity to do. That does not preclude us from holding that in non-defective exercises of a perceptual capacity subjects get into perceptual states that provide indefeasible warrant for 4
McDowell, John.: Perception as a capacity for knowledge. S.31 McDowell, John.: Perception as a capacity for knowledge. S.31 6 McDowell, John.: Perception as a capacity for knowledge. S.37 5
perceptual beliefs."7 If we admit non-veridical perceptions we implicitly admit verdicial ones. A defective use of a capacity is a use of a capacity opposite of what the capacity is per definition supposed to do. And when a subject exercises the capacity as it is supposed to be exercised, i.e. in a non-defective way, the capacity is a warrant for his beliefs. According to McDowell, the very fact that there are cases of illusory perception does not imply that perception, in its non-defective exercise, is not a veridical and indefeasible warrant for our beliefs, insofar as it makes actual states of affairs present to us. By claiming this, McDowell does not intend to deny that perception like all human capacities can be fallible. Rather he wants to defend the veridical power of its non-defective use. Lets take for exemple a basketball player and the capacity to shoot a ball into the basket from the 3point-line. Even the best basketball players do not make every 3point-throw. The capacity they have is however exactly that, to thow the ball in the hoop from the 3point-line. The capacity although sometimes fallible is nontheless a capacity to do a certain thing (in usual non-defective cases). This however does not satisfy the critics. A fallibilist position holds furthermore that the main problem with a perceptual capacity that is defective at times is that the perceiver cannot be in a position to know wether he is deceived or not. The basketball player for exemple knows when his capacity to throw the ball has failed, since in that case the ball doesn not fly through the hoop. A perceiver however who is hallucinating or perceiving an illusion does not know that his capacity is malfunctioning: he might believe to see a green object in front of him although the object is yellow(e.g. due to lighting conditions). In order to discuss this problem, McDowell uses an argument from illusion put forward by Sebastian Rödl: "Whenever I seem to know something (on the basis of perceptual. experience), I might have been fooled. Had I been fooled, I would not have known that I was. I would not have been able to tell my situation apart from one in which I am not fooled. This shows that my grounds do not place me in a position to exclude that I am in such a situation. They do not enable me to exclude that I am fooled. The argument supposes that, had I been fooled, I would have believed the proposition in question on the same grounds on which I believe it now that I am not fooled. This straightforwardly entails that these grounds do not establish the truth of what I believe and therefore do not provide me with knowledge. But when I know something on the ground that, say, I perceive it to be the case, then I would not, had I been fooled, have believed it on this ground, for, had I been fooled, I would not have perceived it to be the case. Hence, when I am not fooled, my grounds exclude that I am fooled: when I perceive how
7
McDowell, John.: Perception as a capacity for knowledge. S.38
things are, I am not fooled with regard to how they are."8 In other words, I can be fooled without knowing that I am fooled. and in that situation my perceptual grounds do not allow me to exclude that I might be fooled and hence do not provide me with knowledge. Nevertheless, those grounds are different from those that I adduce when I know that I know something—that is, in a case in which perception is exercised in a non-defective way. Therefore, from the fact that, when I am fooled, I do not know that I am, it does not follow that, when I am not fooled, I do not know that I am not.9 Rödle emphasizes his argument on the fact that in the two different situtaions we are aducing two different grounds for our beliefs. It is important to discuss Rödls argument a bit more in detail: On a first level, Sanguinetti argues, this argument seems to conclude on a sort of truism: if things are as I perceive them to be then I am warranted to believe that things are as I perceive them to be, because the grounds on which I believe them are veridical. When I misperceive things the grounds on which I believe them are different, they are non-veridical. This seems pretty straightforward and conclusive, yet this does not stir away from the main problem namely: how do we know that the grounds are veridical or non-veridical? Thats why on a second level Rödl is claiming that: "When I know that p as I perceive it to be the case, then I know that I perceive that p. Thus I am in a position to distinguish my situation from any possible situation in which I would be fooled, for, in any such situation, I would not perceive that p, while in the given situation I do." But this cannot hold, because since perception is fallible at times, "knowing how things really are" cannot soleley depend on perception. What makes the distinction between a veridical and non-verdidical ground is not perception but rather the person's "knowing how things are", if you will. For exemple, if I perceive a yellow object I can say: this object is yellow (my grounds are my perceiving it as yellow). If a yellow object is put under certain lighting conditions in which it appears green, I will perceive it as being green and say: this object is green(my grounds are my perceiving it as green). In both cases my grounds are the same (perception) and I would assert that that such and such is the case. Even in the case where i am mistaken I would still hold the object to be green. Thus the grounds adduced do not warrant my belief that I am not fooled and do not allow me to discriminate between defective and non-defective cases. So the grounds on which I distinguish both cases cannot be perceptual but rather as mentined before "my knowing that
8 9
McDowell, John.: Perception as a capacity for knowledge. S.42-43 Sanguinetti, Federico.: Is perception a capacity for knowledge?
the object is yellow".10 This however cannot be given to me from my first person view (perception) but rather by an external source or a third person viewer that knows about the deceptive lighting conditions or lack of them. If Rödl wants the warrant of a true belief to be shifted to perceptual grounds, the ability to distinguish between veridical and non-veridical perceptions is a necessary condition, this however seems impossible. Still, McDowell holds that it would be embarrassing, if due to the mere fact of the fallability of our perception, a perceptual state could never count as making a color visual to us. The mere fact that lightning conditions could be undetectably altered as to make a yellow object appear green is not a reason to say that for all the perceiver knows, the present lightning conditions are unsuitable. For him the sheer fact of fallability does not show that on occasions where the risk of being fooled does not materialize, an observer was not giving expression to knowledge and to knowledge that he knew he had. This however, it seems, needs to be further discussed. In order to do this let us first sum up the two big points in McDowells thesis: (a) perception is always rationally and conceptually shaped and (b) perception provides a veridical and indefeasable warrant for our beliefs about the world. McDowell seems to be right in asserting that perception can tell us how things in the world really are in cases where the perceiver is not fooled. Furthermore it is arguably intuitively true, that perception is "rationality in operation". Nontheless when it comes to the fallibility argument and the indiscriminabilty of veridical and non-verdidical cases, McDowell seems to encounter rather large difficulties. McDowells first thesis, that perception is always already conceptually shaped does not sufficiently support his second thesis, that perception provides a veridical and indefeasible warrant for our beliefs about the world and, hence, provides us with knowledge. Its seems like McDowell's thesis is pretty much build on the credulity that in general perception is a good warrant for our beliefs about the states of the world. Yet the only thing that McDowell seems to really prove is that true perceptions do in fact exist, that a capacity, however fallible, is still a capacity and that from the mere fact that a capacitiy is defective it follows that there are first and foremost cases where the capacity is non-defective. This does however in no way entail that perception is always a conclusive indefeasible warrant. The fact that perception is rationally shaped does not help in discriminating between defective and non-defective exercises of that same capacity. A capacity cannot determine if it is fooled while it is beeing fooled. There would need to be a superordinate capacity that would then be in charge of verifying that first capacity. Only this would make our perceptual capacity a true capacity for knowledge. Of course our perception will be a good warrant in most cases but without this 10
Sanguinetti, Federico.: Is perception a capacity for knowledge?
supervising capacity, perceptional knowledge cannot be anything more than an educated guess. It is viable as far as experience has shown that perception does pretty well in informing us about the states of the world and without trusting it, everyday life would be impossible. This weak interpretation, as Federico Sanguinetti calls it, is probably the only position that can be successfully held. This interpretation claims that perception is in general a capacity whose exercise can lead to true beliefs. If we concede, as we should, that there are cases of veridical perception, we must also conceed that perception is an epistemologically relevant capacity. But if we maintain that a perceiver is capable of knowing in which cases he is exercising his capacity in a non-defective way, we run into problems. We would need to come up with a capacity that lets us discriminate between veridical and nonveridical sates of perception since using the perceptual capapcity itself for this job is erroneous. Since there is no such capacity present to the human mind, one can never selfconsciously determine wether a perceptual state is indeed veridical. Hence our perceptual capacity is a very good guide, which helps us to navigate the world and allows us to form justified believes about the states of the world in most circumstances; yet the utopic thought of it being an allround capacity for truth is however just that, utopic. Although perception can never be a fully safe warrant for knowledge, one should however not fall into the dark pit that is scepticism and completely mistrust perception an deny perceptual knowledge as a whole. As McDowell shows, in usual non-defective cases, where perception as a capacity is exercised as it is intended to by the capacity itself, perception does fully warrant our beliefs about the world. Even if the possibility of non-veridical perceptions cannot be completely excluded, experience and common sense has taught us that perception usually warrants our beliefs about the world. Being tricked by our senses is just an occupational hazard that goes along with being a human being yet completely mistrusting them would be a fatal mistake; a mistake that would result in complete practical absurdity.
Literature: - McDowell, John. Perception as a capacitiy for knowledge. Marquette University Press 2011. Milwaukee, Wisconsin.
- Sanguinetti, Federico. Is perception a capacity for knowledge? A discussion on McDowell's account
of
perceptual
knowledge.
University
of
Padua
/
WWU
Münster.
(https://www.academia.edu/2514794/Is_Perception_a_Capacity_for_Knowledge_A_Discussi on_on_McDowells_Account_of_Perceptual_Knowledge)
- Internet Encyclopedia of Philosophy. (http://www.iep.utm.edu)
