Some of the authors challenge claims that there is a priori knowledge while others produce arguments supporting the existence of such knowledge. The book.
Table of contents
Harman gives examples of how an object could look red all over from one angle and green all over from a slightly different angle, or red all over to one eye but green all over to the other. But the relation between looking red or green and being red or green is complex. These examples do not show that an object could be red all over and green all over at the same time.
Perhaps the thing to say about these examples is that the object is neither red all over nor green all over at the same time. Hence it is not both red and green all over at the same time and so does not contradict 5a. That it just seems obvious that no object could be red and green all over at the same time is an intellectual intuition that provides some justification for believing that proposition. That is compatible with there also being other grounds for believing it true. The most promising account of a priori justification in terms of a nonexperiential source of evidence is one that sees intellectual intuition, rational insight, or apparent rational insight, as providing the relevant a priori evidence with its source being reason, not some special faculty of intuition analogous, say, to sight, which is a source of empirical evidence.
Recently some philosophers have thought that a person can be justified in believing, or accepting, a proposition without having any evidence to support it, and so even if there is no nonexperiential source of evidence for that belief, or acceptance. Timothy Williamson has argued that certain acquired skills can provide justification for believing a proposition for which the person does not have evidence. For instance, a person might be justified in believing that two marks nine inches apart are further apart than the front and back legs of an ant, even though he is not now looking at an ant, and so lacks visual evidence to support his belief Willamson b: The proposal is that it is his skill at making comparisons of length, not his evidence, that supports his judgment.
Even if true, it seems that we can distinguish empirically based from understanding based skills. On some accounts of a priori justification, namely those that hold that a priori justification is justification independent of empirical evidence , general foundationalism would imply that all of a person's beliefs are prima facie or weakly a priori justified since that justification would stem merely from the fact that the person believes them, not from any empirical evidence that supports them. On this view, and contrary to initial appearances, there is really no difference in the way the propositions at the start of this essay are prima facie justified since they are all weakly a priori justified if you accept them.
If some are all things considered less justified than others, it is because of the relationships between them. Coherence considerations account for all things considered justification and are what upsets initially equal prima facie justification. Further, this view has the implication that you could be prima facie justified in believing extremely bizarre propositions, say, beliefs about what happens on the planet Gliese d Russell Insofar as you have no defeating evidence, you could even be all things considered justified in believing all those things about the Glieseans, despite having no evidence to support your beliefs.
So he adds the requirement that an a priori justified belief cannot be empirically defeasible Field But that requirement seems too stringent.
- Bestselling Series.
- Capabilities!
- Dirty Little Secrets.
- Did You Know??
Suppose I have an intuition that, necessarily, if P , then Q for some specific P and Q say, that necessarily, if a person has a justified true belief, he has knowledge. Then it is possible for me to be a priori justified in believing that conditional is true. However, if I think of a possible example where P obtains but Q does not, that a priori justification will be defeated. But an actual example known empirically to obtain of a P that is not a Q is a possible example.
A student once described such an example to me. She did not want her mother to know she had gotten a tattoo.
No Place for the A Priori *
So she threw away the receipt. But her mom found her boyfriend's receipt with no name on it in the glove compartment of her car and, on that good evidence, concluded that she had gotten a tattoo! Her mom had a justified true belief, but not knowledge, that my student had gotten a tattoo. So actual examples known empirically to obtain can defeat a priori justification. Further, I might be a priori justified in believing the conclusion of some mathematical proof I have constructed, but that justification can be defeated by the testimony of excellent mathematicians who tell me that the proof is unsound.
Since testimony is an empirical source, this is another example of how a priori justification can be defeated by empirical evidence. Thanks to an anonymous referee for suggesting this example to me. It is a mistake to think that a priori justification cannot be empirically defeasible. Of course, if Field allows that a priori justification can be empirically defeated, he faces the same problem as Harman in drawing the boundaries of a priori justification too broadly.
A final attempt at offering a view of a priori justification that does not rest on nonexperiential evidence holds that we are entitled to accept certain propositions on no evidence and that entitlement on no grounds evidence is what a priori justification amounts to. To be entitled to accept, or trust, some presupposition is for it to be rational to accept or trust it, though this is supposed to be different from being justified in believing it. Crispin Wright proposes that the laws of logic and the presupposition that I am not now in the midst of a coherent and continuing dream, not now a brain-in-a vat, etc.
Of course, the gains and losses must not be pragmatic gains and losses gains and losses in happiness, prestige, accomplishments, wealth and the like. Otherwise all that would follow is that it is practically rational to accept the presuppositions. The gains and losses must be epistemic say, having to do with truth, or probable truth, or with evidence , for Wright wants the rational acceptance of such presuppositions to be an answer to the skeptic. But even if we accept that the gains and losses must be epistemic, there seem to be counterexamples to Wright's view.
Given i — iii above, Wright should count the view that there are no miracles as a rational presupposition of science because i science could not be pursued if it assumed that some phenomena might have supernatural causes then why continue to look for natural causes instead of positing some supernatural cause of some phenomena? But even if within the cognitive project of science we are entitled to accept that there are no miracles, it does not seem that we are entitled a priori to accept it.
Any such entitlement would seem to rest on inductive grounds: So probably in the future they will be explained in naturalistic terms, too. Counterexamples aside, some have questioned whether the project-relative rationality of a presupposition determined by the epistemic value of the consequences of accepting it is enough to make it rational to accept that presupposition Jenkins For instance, when conducting certain inquiries, it might be rational relative to some project or kind of inquiry to accept that the world is a pretty orderly place, yet not epistemically rational to accept the presupposition itself maybe we should suspend judgment about that until we go look at the world.
This objection parallels a standard objection to rule egoism in the practical realm in which critics point out that it may be in a person's rational self-interest to accept a set of rules analogues to Wright's presuppositions for action but not in her rational self-interest always to act in accordance with some rule in that set the analogue to its not always being rational to believe in accordance with those presuppositions. Sometimes Wright does not seem to appeal to a dominance argument to make his case that it is rational to accept a presupposition. But that seems to allow a presupposition based on a whim that does not rest on any more secure presuppositions could be justified, and even a priori justified.
In summary, it seems that accounts of a priori justification that do not hold that it rests on evidence provided by a nonexperiental source are in danger of counting certain beliefs or acceptances as a priori justified that, intuitively, do not seem to be. The attempt by Field to narrow the circle rests on a false assumption, namely, that a priori justification cannot be defeated by empirical evidence. Since a correct account of intuition can explain why we are justified in accepting the laws of logic and, I'd add, of induction and inference to the best explanation , these cases do not represent problems for rational intuitionism.
It's true that rational intuitionism cannot explain why we are a priori justified in accepting the proposition that we are not now in the midst of a persistent coherent dream or in accepting that there are no miracles, but that is not a problem. We can be a posteriori justified in believing these things even if we are not a priori justified in accepting them. I will now turn to considerations that seem to count for the view that intellectual intuitions are evidence for the propositions that are their objects.
There are four arguments for the view that intuitions can provide evidence. The first argument sounds circular because it starts with examples like Sheep and claims that it is intuitively obvious that the intuition that the person does not know there are sheep in the field is evidence that he does not know. Here a second-order intuition about the evidential weight of first-order intuitions is invoked.
This approach goes on to defend the appeal to intuitions on the grounds that according to our standard justificatory practice intuitions as well as experiences visual, tactile, auditory, olfactory, taste provide evidence. It then challenges those who want to eliminate intuitions as a basic source of evidence to say why this is legitimate but, say, eliminating all but visual experiences as basic sources of evidence is not Bealer a,b, , The second sort of argument rests on the idea of concept possession Bealer Concept possession guarantees the reliability of concrete case intuitions.
A person would not possess the concept of a square if he often applied the term to circles and ovals or did not apply it to squares whose sides were over three hundred feet long. In general, to have the concept of an X , a person must be reliable though not infallible in his application of the term X to, and only to, things that are X. However, even if concept possession requires reliability, this does not show that concept possession guarantees justification since reliability does not.
There are lots of cases where, intuitively, reliability is not sufficient for justification. Keith Lehrer's famous Truetemp case is one of them Lehrer On the basis of his first clairvoyant experience, Norman finds that he cannot help believing something say, that the President is in New York , but, intuitively, he is not justified, and does not know, what he can't help believing. This example shows that reliability by itself is not sufficient for either justification or knowledge. So there is no reason to think that even if concept possession guarantees reliability, it guarantees justification.
George Bealer thinks that a certain sort of concept possession what he calls determinate possession guarantees necessary reliability, and that, in turn, does guarantee justification. Bealer thinks that the following is false: However, he thinks the following is true: In addition, he thinks that a certain sort of concept possession necessarily makes our judgments about the applicability of concepts to concrete cases reliable. A third attempt to show that intuitions provide evidence relies on what we might call an outside source of evidence.
We might rely on the standard five senses to confirm the reliability of some sixth sense say, echo-location , and then later rely on that sixth sense alone to provide justification. Similarly, suppose the best explanation of my failures, and the failures of others, to find counterexamples to propositions that seem intuitively obvious is that those propositions are necessary truths. That could give me reason to believe that having an intuition that P is good evidence that P is necessarily true.
In turn, that could justify me in later relying on intuition alone to provide justification for the proposition that is its object. A fourth approach offers ideal conditions under which rational insights, or intellectual intuitions, provide evidence requiring ideal conditions is not unique to this approach as Bealer b and and Peacocke also require them in their defense of intuition based on concept possession. BonJour argues that a priori justification that rests on rational intuitions, or insights, does not require what he calls a metajustification for those intuitions to provide justification, that is, does not require reasons, or an argument, to show that beliefs based on those intuitions are likely to be true.
In this respect, intuition is like introspection and unlike perception, premonitions about the future, and clairvoyance if it exists. BonJour seems to think that a principle something like the following is true: Here though not elsewhere he seems to assume that beliefs that are likely to be true are justified, so J implies that subjects that meet the conditions specified in its antecedent are justified in believing p.
Given BonJour's objections to reliability as being sufficient for justification, he probably means that a belief that satisfies conditions i — iii is epistemically likely to be true, where that sort of likelihood does not imply objective likelihood and can be present even in demon and Matrix worlds. However, BonJour does not believe that a person must be justified in believing J or some similar principle in order to be justified in believing some proposition on the basis of an intuition that, as a matter of fact, satisfies conditions i — iii.
He seems to think that if J could be justified, it would be either by empirical evidence or by intuition. If empirically justified, he thinks that rational intuition would not then provide a priori evidence. On the other hand, justification by intuition would be circular since J is supposed to give the conditions for intuitive justification. So it seems that there is no way to justify a principle that states the conditions sufficient for a priori justification: So if justification via rational intuitions is possible, it cannot require a metajustification of some principle like J.
Still, according to BonJour, if J is in fact true and rational intuitions satisfy the antecedent conditions of it, they will provide a priori justification. After all, J 's truth is not affected by any sort of justification, and it is a statement of when intuitions provide a priori justification. All that an empirical justification will provide are reasons to think J true, not reason to think intuitions do not provide a priori justification. Of course, if you thought that empirical justification presupposes a priori justification, and that that sort of justification presupposes J , there could be no non-circular justification of J.
It might seem obviously true to me that I am going to win the next lottery or that there are more females than males on Gliese d , where that seeming is not the result of some inference. Several philosophers appeal to the understanding in their accounts of a priori justification: Bealer in all his essays; BonJour ; Jackson ; Peacocke Then he would have a non-inferential belief in that proposition based solely on his understanding it, and so on this account would have an intuition, and be justified in believing, that it is true.
An alternative defense of the evidential force of intellectual intuitions would say that when a person is a priori justified in believing some proposition it is because she possesses the relevant concepts expressed in that proposition. Possessing those concepts, and understanding the proposition that contains them, explains why the person is justified in accepting that proposition even if it does not cause them to hold it. The concept possession, or understanding, of the person with the odd psychology who believes contingent propositions once he understands what they are asserting may cause him to believe those propositions, but it does not explain why he is justified in believing them.
Strictly speaking, a person's acceptance of some proposition is based solely on her understanding it only if that understanding both causes and justifies her accepting that proposition. Hence, strictly speaking, the beliefs or acceptances of the person with the odd psychology would not be based solely on his understanding. A new branch of philosophy called experimental philosophy X-phi for short has studied the intuitive judgments of people often students when presented with well-known examples in epistemology and ethics.
They ask these people often from different ethnic, cultural, economic, and educational backgrounds whether someone in a hypothetical scenario knows, or only believes, that some proposition is true, say, in Sheep whether the person knows, or only believes, that there are sheep in the field. In ethics they may present the subjects with a case and ask them if it is wrong, or not wrong, to do what is described. In a case often called Transplant , five innocent people are desperately in need of certain vital organs, and the only way to save them is to cut up some innocent person and distribute his organs to the five transplant surgery has been perfected and our potential donor is a perfect match to all five.
Experimental philosophers will ask their subjects whether it is wrong, or not wrong, to cut up the one to save the five, and then record their intuitive judgments. In another case often called Trolley , a runaway trolley is on track A and headed for five innocent people who are trapped on that track. All person S can do to avoid this is to turn the trolley down track B where one innocent person is trapped. If S does nothing, five will die; if he throws the switch via a remote device, the one on track B will be killed.
Experimental philosophers ask whether it would be wrong, or not wrong, for S to throw the switch. They record the data, which they take to be intuitive judgments on the cases, and note differences in the responses, say, between different ethnic or economic groups. Maybe the intuitions that are studied by experimental philosophers do not provide evidence but intuitions understood as being based solely on the understanding do see sec. So how could intuition be a source of evidence given that there is so much disagreement in intuitions and no way to discount some and privilege others?
Top Authors
Suppose we had the ability to see clouds in the sky but nothing else, nothing on earth and nothing solid in the sky. Of course, the answer to that question depends on what is the best explanation of the disagreement. Similar sorts of consideration need to be looked at when it comes to the differing intuitions observed by experimental philosophers. We need to know what the best explanation of the disagreements is before concluding anything about the evidential status of intuitions. Suppose we grant for the sake of argument that experimental philosophers have observed disagreement in intuitions in the relevant sense, that is, where an intuition is understood to be an intellectual seeming founded solely on the understanding of concepts.
Before concluding that intuitions cannot provide evidence, we must ask why they disagree. In so-called Truetemp cases, a person is able to correctly tell the ambient temperature because, unbeknownst to him, his brain has been affected by a blow to the head. But do they make different background assumptions? Do more Westerners than East Asians assume that the person has confirmed his ability to correctly tell the temperature?
If so, there may not be a fundamental difference in their intuitions because the members of the two groups of subjects will not be considering precisely the same case. Until we investigate further into whether different background assumptions are being made, we should suspend judgment about what to conclude from the different responses.
This difference can remain even when subjects are first screened to make sure that they have at least minimal competence with the relevant concept say, of knowledge. Until there is some reason to think that the differences in intuition are not the result of people employing different concepts and are not due to different levels of understanding of the concepts, the experimental philosophers should suspend judgment on whether intuitions, properly understood, can provide evidence.
The crucial role of the understanding in grounding intuitions, when properly understood, might make one think that the only relevant data regarding different intuitions are different intuitions among professional philosophers who probably share equally high levels of understanding of the concepts. Of course, if differences were found here it might be because some of these philosophers have a vested interest in defending a view that to others has counterintuitive implications.
On the other hand, if few differences were found, this might be due to the education of graduate students and the culture of professional philosophy. Perhaps graduate students are taught that certain Gettier cases defeat the justified true belief account of knowledge and, for the most part, students with contrary intuitions do not make it through graduate school!
Perhaps a person's colleagues look down upon, or dismiss, someone who does not share intuitions that are widespread in the discipline. Insofar as there are several competing explanations for disagreement in intuitions properly understood , and one is not obviously better than the others, we should suspend judgment about the evidential weight of intuitions, just as we should suspend judgment on whether a coyote or a dog has left certain paw prints in an area where it is not known whether it's more likely to be dog than coyote paw prints or vice versa.
Some initial differences in intuitions e. Other differences have been found to obtain only when two or more cases are considered together but not when considered separately. Whether the data support the view that intuitions provide evidence, or support the opposite, depends on what the best explanation of those differences is. I am indebted to Beebe for updating me on the current state of experimental philosophy. His essay also raises some of the same objections and problems that I discuss here. And if the data collected only concerns intuitions understood in a loose, commonsensical way, then the conclusions reached will not be relevant to the epistemic claim that intuitions, properly understood , have evidential weight, at least when they are the intuitions of people with a deep understanding of the relevant concepts where those people are not influenced by bias and prejudice that is, where the intuitions are produced under ideal conditions.
There are other objections to the reliance on intuitions in philosophy that do not call into question their reliability. They call into question their relevance. Kornblith thinks that intuitions can help direct us to the appropriate objects, or phenomena, of investigation but not much more. For instance, we have an intuition that knowledge is not a type of furniture so we should not start our empirical investigation into the essential nature of knowledge by looking at furniture Kornblith , , So the fact that we cannot have a priori knowledge of propositions that contain natural kind terms is no reason to think that we cannot have a priori knowledge, period.
The problem with this empirical approach to philosophy is that the terms that philosophy is interested in do not seem to be natural kind terms Feldman However, the properties relevant to knowledge are part of the essence of knowledge e. Similar remarks apply to other terms that have been of interest to philosophers: Another approach that discounts the role of intuitions in philosophy, especially in epistemology, is pragmatic.
The idea is to first determine what epistemic goals we want principles to serve, and then to discover empirically which epistemic principles, if adhered to, will best serve those goals Weinberg For instance, your goal might be to have lots of true beliefs or, alternatively, to have few false ones. Or your goal might be to have beliefs that make you happy. Probably the best set of rules to follow to obtain lots of true beliefs will be different from, and more lenient than, the best set of rules to follow to avoid having false beliefs, and it is probable that those sets of rules will be different from the set of rules you should adopt if you are interested in having beliefs that make you happy.
It's reasonable to think that intuitions will have to be appealed to in determining what the appropriate epistemic goal is, and it is probably different from any of the three just considered. For instance, Lehrer Intuition must be relied on to determine that the goal of making a person happy is epistemically irrelevant, that is, that a belief cannot be epistemically justified simply because a person's holding it makes him happy. It's just as counterintuitive to think that the appropriate epistemic goal is to have many true beliefs, for that goal could be satisfied by memorizing all the phone numbers of people in the phone book.
The pragmatic approach that I have sketched seems doomed at the outset: Another problem that faces the pragmatic approach is that whether we are empirically justified in believing some epistemic principle, P 1, say, that you should believe a proposition for which you have the most confirming instances serves the relevant epistemic goal, G 1, say, having many true beliefs depends on whether some other epistemic principle, Q 1, implies that we are say, a principle that says that adherence to the epistemic principles you accept, like P 1, are more likely to lead to truth if you have consulted a crystal ball and the crystal ball says that adherence to them is more likely to lead to truth.
But what justifies us in believing that these other principles like Q 1 are true? On the pragmatic approach, it must depend on whether adherence to them serves relevant epistemic goals G 1 or some other goals, G 2, G 3, etc. Well, they either do or do not, but we want to know whether we are justified in thinking they serve the relevant goals. That must be determined by some third set of epistemic principles, R 1 maybe R 1 says that whether adherence to Q 1 is likely to lead to truth depends on whether the reading of tea leaves says it will.
At each step we must presuppose certain epistemic principles in order to assess whether we are justified in believing that certain lower level principles serve the relevant epistemic goals. We could bring this regress to a halt if we could appeal to epistemic intuitions to assess the presupposed principles, but the pragmatic approach disallows such an appeal.
A different sort of objection to intuitions as a source of a priori evidence assumes that a source of justification must be capable of being calibrated to determine whether it is accurate Cummins What we see through a telescope justifies us in believing that the moon has mountains because we have looked through telescopes at distant mountains on earth and then gone to them and discovered that the telescopes presented an accurate picture of the mountains. But what, the objection goes, can intuitions be checked against?
But that is like checking a crystal ball against itself. BonJour has argued that many errors involving apparent rational insights intuitions can be corrected internally by further reflection, or by appealing to coherence BonJour Others have replied that neither perception nor memory Goldman In reply to this sort of response, critics of intuition-based views of a priori justification have said that at least different types of perception can be checked against each other, say, vision against touch Weatherson The critics of intuition add that while we can distinguish circumstances where, say, vision is unreliable say, where the lighting conditions, or the person's eyesight, are bad; when we are in a desert where optical illusions occur; when we are hallucinating, etc.
However, the latter does not seem true. I can tell when I do not have a very firm grasp of some concept say, of poignancy , and sometimes I know that someone has thought about some concept as long and as hard as I have is my epistemic peer , has intuitions opposed to mine, and I cannot explain his intuitions away. In those circumstances, the evidential weight of my intuitions is at least diminished and may be reduced to zero. Someone might respond that this only shows that we can know under what conditions intuitions are not reliable, but we do not know under what conditions they are reliable.
Sometimes we can check one sense modality against another. But suppose we could check a Ouija board against a crystal ball, and they always agreed. That need not give us reason to believe that either is a reliable source of truth. Perhaps some agent or force is causing the two sources to agree even though what they agree on is false; perhaps there is a common cause that explains the relevant agreement.
Suppose, for instance, over a good stretch of time both sources agreed on what the weather has been like on Gliese d. I think we should suspend judgment about whether they are accurate in their weather reports if we do not have any sensory input that bears on the issue. So agreement between, say, vision and touch may not justify us in believing what they agree upon. Again, it depends on what the best explanation of agreement is whether it be agreement between the Ouija board and the crystal ball, sight and touch, or between the intuitions of different people.
Further, why should we accept the claim that a source of justification must be capable of being tested for its reliability? Even if we cannot test whether we are in The Matrix or some Cartesian demon world, it does not seem to matter as regards justification. Perhaps that is because justification is a matter of being epistemically responsible to the evidence empirical or a priori , not a matter of reliability, though just what epistemic responsibility amounts to is a difficult question.
Finally, it may be true that intuitions do not have as much evidential weight as perceptions. It seems that coherence considerations can override intuitions more easily than they can override perceptions.
What Place for the a Priori? - Google Книги
A person who initially has the intuition that someone who holds a losing ticket in a large lottery does not know that he does may justifiably give up that intuition on the basis of some theory of knowledge that implies that the person does know that he holds a losing ticket. However, someone who sees a humming bird or bumblebee fly should not give up his belief that he saw them fly just because some physical theory he accepts implies that these creatures cannot fly and so he could not have seen what he thinks he saw.
Suppose, for the sake of argument, we grant that intuitions properly understood and had under ideal conditions by people with a deep understanding of the relevant concepts can justify certain propositions. But can they justify propositions that are about the world?
Carrie Jenkins has argued that they can insofar as the concepts that play a role in a priori justification have been shaped by experience. She thinks that for knowledge not justification our concepts must be grounded. By this she means that they must accurately and non-accidentally represent the world. So the concept table can be grounded for a person in a world where there are tables but not for a brain-in-a-vat BIV Jenkins a: So a BIV can have a justified, though not a grounded concept, about things existing in the external world. Jenkins's argument for thinking our concepts are grounded is that our basic concepts are useful, but if they did not fit the world weren't grounded even though they are founded on sensory input, their usefulness would be a miracle.
But since we should not believe in miracles, those concepts must fit the world. The best explanation of the usefulness of our concepts is that they accurately represent features of the world that produce our sensory inputs that allow us to navigate successfully in the world. She thinks that this No-Miracles Argument shows that it is reasonable to think that our concepts or groups of concepts mirror the world's structure Jenkins a: If we have justified concepts, we can examine them to see what they involve and then have a priori justification for believing that certain propositions that involve them are true of the world.
So, on her view, we and BIVs could be a priori justified in thinking, merely on the basis of examining our concepts , that all vixens are female and that there are or at least were vixens, and that all bachelors are unmarried and that there are or at least were bachelors. If we have grounded concepts, we but not BIVs can have a priori knowledge that all of these propositions are true.
Jenkins often appeals to an analogy with maps when discussing the accuracy of concepts. Accurate maps accurately represent the location of various things in the world and their location relative to each other how they fit together. Accurate concepts represent how things are in the world and how they relate to each other how they fit together. But maps do not represent how things are in every possible world in some possible world San Francisco is not south of Seattle: This-worldly grounding does not seem able to explain how we can have a priori justification and knowledge, and Jenkins's account of a priori justification and knowledge that is built on her notion of grounded concepts has the counterintuitive result that we can be a priori justified in believing, and can even know, things about the world say, that vixens and bachelors exist or existed simply on the basis of examining our concepts.
It's more plausible to think that we have a priori justification for believing that if someone is a bachelor, he is unmarried and if some animal is a vixen, it is female and a posteriori empirical justification for thinking there are bachelors and vixens. Together these two propositions imply that all bachelors are unmarried all vixens are female and that there are bachelors vixens. But since the justification of that conjunction rests in part on an empirical premise, its justification is empirical, not a priori.
Further, we could know the relevant conditionals a priori even if we were BIVs , provided we adequately understood the concepts involved. We do not need to have accurate concepts, ones that fit how the world actually is, in order to have knowledge of the relevant conditionals. We could have conceptual knowledge founded on the relation of concepts regardless of whether the concepts were implanted in us by some scientists or supercomputers as in the film The Matrix or God, which shows that not even enabling experience is necessary for a priori knowledge though it is compatible with it , let alone having grounded concepts which require sensory inputs caused by things in the real world.
Further, there seem to be actual examples of propositions that are not founded on justified or grounded concepts that we can be a priori justified in believing and even know. If the propositions that are supposed to be the objects of a priori justification and knowledge are really counterfactual conditionals, then our justification for believing them could rest on empirical evidence Williamson b. We can be justified in believing, and even know, that if the earth stopped spinning half of it would be boiling hot and half freezing cold on the basis of knowing about what makes objects hot and cold, that the sun is hot, that the earth does not produce its own heat, etc.
On the view we are considering, we know that if someone is a bachelor, he is unmarried on similar empirical grounds because that conditional should really be understood as a counterfactual conditional of the form: That counterfactual conditional is true if, and only if, in the nearest possible world where the antecedent is true, so is the consequent.
It is widely, though not universally, held that knowledge requires justified true belief. If so, it is plausible to understand a priori knowledge as knowledge founded on a priori justification, that is, as a priori justified true belief, plus some anti-luck conditions to handle Gettier cases. Whatever is the correct account of a priori justification could then be used to fill out the details of the account of a priori knowledge. The chief rival to accounts of a priori knowledge that require a priori justification is knowledge reliabilism that holds that someone knows something if, and only if, they have a true belief that has been reliably produced, that is, produced by a cognitive mechanism, faculty, or process that, for the most part, yields true beliefs, and some would add that knowledge is not ultimately defeated by other considerations there are no undefeated defeaters.
Presumably, knowledge reliabilists would say that a person knows some proposition a priori insofar as he has a true belief, his belief that it is true has been produced solely by his understanding of the relevant concepts involved, that understanding is a reliable source of true beliefs, and that knowledge is not defeated by any undefeated defeaters. If there are problems with this reliabilist conception of knowledge, they are similar to the problems faced by reliabilist conceptions of justification.
One of those problems is that a person's beliefs can be formed by a reliable cognitive mechanism without that person's having any confirmation, or evidence, that that mechanism is reliable. Let Truenorth be a person who knows what direction is north, what south, what east, and what west even after being spun around several times with his eyes closed.
Perhaps Truenorth has some internal compass like the ones found in migratory birds. However, assume that Truenorth has no reason to think that his beliefs about compass directions are accurate; he has never received confirmation of their accuracy, neither from the testimony of others or by checking things out himself. Nevertheless, he is confident that his beliefs about what is north, etc. Assume, also, that he has no reason to think that others in his society have, nor that they lack, his directional ability.
In general, assume that there are no undefeated defeaters of Truenorth's beliefs about what is north, etc. Intuitively, it seems that if he believes that some direction he points to is north, he is not justified in believing, nor does he know, that it is north, even if what he believes is true.
Available on
If this is a problem for reliabilists about empirical justification and knowledge, it is also a problem for reliabilists when it comes to a priori justification and knowledge. Traditional attempts at explanation that appeal to analyticity fail in two ways. They rest on an unexplained acceptance of logical truths and on the mistaken view that competence with a concept is sufficient for knowledge about it.
A consideration of the contemporary views of Peacocke, Bealer, and BonJour helps to bring out the obscurity. Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter. Please, subscribe or login to access full text content. To troubleshoot, please check our FAQs , and if you can't find the answer there, please contact us. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use for details see www.
University Press Scholarship Online. Publications Pages Publications Pages. Search my Subject Specializations: Classical, Early, and Medieval Plays and Playwrights: Classical, Early, and Medieval Poetry and Poets: Classical, Early, and Medieval Prose and Writers: