Relative KnowledgeMembership Status
Registration Date: Mar 10, 2002Usergroup: Members
Articles Submitted: 3 articles
Total Comments: 13059 (4 per day)
Total Time Online: 72 days, 1 hour and 16 minutes
Last seen: 05/16/12 - 12:23:43
Biographical Information
Location: Northern CaliforniaBirthdate: July 5, 1980 (31 years old)
Interests: php, writing, philosophy, BBC7
Biography: Not worth writing at present time.
Contact Information
Guestbook: View or sign guestbook (36)Submission History
Most recent articles:Knowledge of Gold
Modeling Time: Presentism and the Block Universe
Reapproaching Idealism
Utilitarianism and Impartiality
Reality, Existence, and the Atom
Find all articles submitted by Paul
By Paul
05/20/03 (Edited 02/27/12)
The question "How do I determine what I know?" is perhaps the most central question to epistemology. The answer to the question depends heavily on a conception of what knowledge is, and on a conception of what perspectives it's possible for a person to have to get at that knowledge. Foundationalist systems of knowledge are appealing because they make the process of deducing one truth from another logical and easy to follow, but they run into major problems attempting to determine what a properly basic belief could be. The requirement that at some point supporting the whole structure there must be a belief which does not need justification presents a paradox -- it's the mystery of the belief which justifies everything else, but has no justification. Clearly, there's something wrong with such a picture... yet since there's also much about foundationalism which seems right or at least useful, scrapping foundationalism altogether would seem unwarranted. Instead, a new approach must be devised for defining a foundation.
The idea of knowledge can be divided into two related ideas of knowledge: absolute knowledge and relative knowledge. By distinguishing between absolute and relative knowledge, and then examining how the base of relative knowledge ties into perspective, we can begin to clear up problems and make epistemology more clear. Among the consequences of this analysis of knowledge is that Gettier counterexamples are rendered unimportant, and the role of externalism can be greatly restricted and replaced by more useful methods of determining what is known.
Absolute and Relative Knowledge
Justification is what converts a belief into knowledge. The level of the justification determines whether in common language we call it a belief, a justified belief, or knowledge. If a person walks up and says "it's going to start raining in a moment," this will be taken as simply the person's belief at first. Once the person gives some justification, such as "I heard it from the weather man" or "those clouds up there are really dark" it becomes justified belief. For it to become knowledge, the person must have much stronger justification (since dark clouds don't always bring rain and weather men are wrong more often than not). For example, if he or she can show you a radar map demonstrating that it's currently raining in all directions around your location and the wind is blowing the rain in towards you, that would convert the belief into knowledge. Or, for even stronger knowledge, the person could be rocketing down from the sky where they've just seen the raindrops falling which haven't yet reached the ground but will get here in a second. At this point, everyone using common language would say that the person knows that it's going to start raining in a moment. It's possible they could be wrong, since it cannot be proven that a sudden 300 MPH gust of wind won't blow everything around such that the rain circles around but never touches your particular spot on the ground, but the possible incorrectness does not and should not prevent people from labeling it as knowledge. It's here where distinguishing relative knowledge from absolute knowledge provides a better understanding of what a person who says "I know it's going to start raining in a moment" means.
Absolute knowledge is the ability to justify every justification in terms of every other justification. Hume's problem of induction shows that absolute knowledge is impossible for non-omniscient beings -- it cannot be guaranteed that an unobserved fact in the universe will not contradict what you think you know, and if you're non-omniscient you can't observe all facts. Even if everything you've observed tells you that the raindrop currently a meter above the ground will hit the ground, you cannot guarantee with absolute certainty that the Earth won't be vaporized by a Vogon constructor fleet before the raindrop can fall. Thus, absolute knowledge becomes the realm of gods only. It's also worth noting how the concept of absolute knowledge came about... it didn't arise from someone knowing something absolutely, but rather it arose from someone feeling their knowledge of something was strongly justified and extrapolating to infinite justification based on that. Absolute knowledge, then, is just an idea extrapolated from relative knowledge that includes infinite justification. It serves as a limit which can be approached, but not reached... yet much like limits in mathematics, it remains a useful concept in that the idea of it helps in understanding that which may approach it.
Non-absolute knowledge, meaning the knowledge which is attainable, is relative knowledge. Relative knowledge involves a group of base assumptions as a foundation, with a chain of justifications on top of that base to support the belief and convert it to knowledge. This is the type of knowledge which we normally talk about, the type that someone means when they point to a falling rain drop above them and say "I know it's raining." Relative knowledge, as the name implies, cannot operate independent from a base of assumptions. You can justify your belief that it's raining by looking out the window and seeing water falling from the sky, but that chain of justification (a very short one in this case, simply perception plus an idea of the definition of rain and what rain usually looks like) only works relative to a base that includes assumptions such as the assumption that there isn't a team of skeptics on your roof with a sprinkler and the assumption that you didn't just take a hallucinogenic drug. However, since these two claims are not likely to be disputed in most circumstances, they can safely form a base and people will readily describe this situation as knowledge. There are a great many beliefs which people are compelled towards, perhaps by natural selection's way of developing the brain or simply the way perception works or the way experience teaches... it's not hard to find common ground between people when it comes to the very central parts of a base group, even though different people often have different assumptions around the edges.
The key with relative knowledge is that the base of assumptions must be small enough that at least most of the listeners will not dispute the validity of the foundations which the justification of the knowledge claim rests on top of. The base should not include items that will be debated -- the evidence on top of the base which makes up the justification of the belief is what should be open for inspection and debate, so anything not obvious must go in the justification rather than the base. The size of base this requires depends on your situation and conception of what's obvious. One can justify "there is a real hand in front of my face" relative to a base which includes "things have reality independent of perception." This relative knowledge will be accepted and adopted by people in most places, much to the satisfaction of G.E. Moore, but will earn only a laugh on the island of absolute idealists where the ghost of Berekely roams and teaches all children from infancy. The claim is still relative knowledge to the speaker either way, but the two situations -- a materialist vs. idealist audience -- determine if the relative knowledge will be considered by the listeners to be knowledge relative to their own base. The presuppositions (base assumptions) of the audience determine what can be built on to create knowledge for them, and in speaking the desire is usually to translate knowledge for the self into knowledge for the listeners.
In certain enterprises like philosophy, the audience will accept very little and the base needs to be made as small as possible... to know that q in a room of philosophers, who will debate the existence of p forever, is in fact much harder than to know that q in a room of regular people who assume p and can thus simply be shown that p implies q. In general, the smarter the person is (or intended audience is) the less is required to be in the base. For example, while virtually everyone else has the classical conception of time within their base, Albert Einstein didn't need to have our concept of time in his base. Of course, it's safe to say he began with our concept of time but removed it later due to evidence he came across which contradicted it -- encountering evidence is how all people change the nature and size of their base over time, but it only works well if one knows how to replace the gap which the removed item will leave in every future attempt at justification of belief. After the reduction Einstein was able to build up from lower levels using spatial dimensions to replace time, which made the previous reduction of the base prove useful. Of course, note that there are plenty of base assumptions Einstein himself had to stick to since he had no way of replacing them. Not the least of these assumptions is the validity of mathematics -- as Godel's incompleteness theorem shows, math can't be proven logical and so must be assumed in order to be used.
The extreme case of an audience who requires a small base is the global skeptic, who will deny any assertion and defend agnoilogy to what may or may not be the grave, if there's such a thing as a life to begin with. The skeptic, however, fails all the tests which those like Einstein pass. The skeptic doesn't need intelligence and so simply chooses to lose the ability to function by reducing the size of the base below what they're capable of dealing with. The global skeptic tears down recklessly, without having any interest in building things back up. (The local skeptic, on the other hand, usually has more interest in maintaining a coherent scheme and questions for scientific reasons.) A global skeptic who truly does contract their base of knowledge into nothing is simply a dysfunctional person, but there is at least one advantage an exploration of global skepticism has in understanding, which although not useful to the average person can benefit a philosopher: the extreme skeptic has a better conception of what sorts of things are in our group of base assumptions. A non-skeptic isn't aware of exactly what they've assumed -- the base assumptions are formed by a coherentist process, driven by the person's experiences in life and things they've been taught to accept along with personal "intuition." When I say "the moon exists" I don't have to consciously realize that I've assumed that NASA isn't actually involved in a vast conspiracy placing a flat slice of cheese in space which is carved to resemble a moon. I'd have to entertain such a proposition in order to know that I've assumed it, and my coherentist web of beliefs would not normally compel me to consider such a scenario... I would only consider it if some sort of evidence came up which I could not explain without challenging my assumption. Due to this, it's the global skeptic who has a fuller idea of some of the specific beliefs and some of the general categories of beliefs which are among people's base assumptions. To most people, though, this doesn't matter until evidence counters the assumption -- without counter-evidence, there's little reason to revise a base assumption.
The way in which this base of relative knowledge is selected does, as explained, depend on other people and social context. This makes it important to consider the perspective involved, in order to gain a much clearer sense of exactly how the size of the base is determined.
Perspective
Normally, philosophers speak of the distinction between the objective and subjective perspectives. The objective perspective is what exists independent of perception and experience, which presumably gives rise to the subjective. The subjective perspective is all experience, and contains all knowledge which the self possesses. Considering only these two perspectives, however, doesn't account for everything we can speak of and would lead to the common misinterpretation of objective reality as being what's common to all people's perceptions. In actuality, the objective is independent of all perception -- to speak of something that's common to perception, or to speak in any other third person manner, is a distinctly different idea. The fact that something is common across subjective realities can just as easily point to commonalities in the perceptual apparatus of the observers being considered as it could to actual commonalities in objective reality. (An object is green to all humans perhaps, but who's to say that the sonar-like perception that a bat has of the object isn't more true to objective reality than any color?)
The third person subjective is a perspective we often take. The third person is a logical partition of the subjective, but it's an important distinction in that the third person subjective world is one which we often consider ourselves to live in when we think of what the world would be like without the self. Third person is about what the "normal" observer external to the situation being considered would perceive... it's a prediction of what a generic person who you know nothing about would perceive a certain situation. This isn't the same as the regular subjective, because it can differ from what you yourself perceive. For example, suppose you've been blind for several years and now you're standing near a building which someone who you trust has told you is painted white. You wouldn't say you perceive that the building is white, because you're blind. You would, however, say that the building is white -- and you would think of the building as actually being white in your mind. This is because your concept of the 3rd person subjective is that the average person would perceive the building as white. Subjectively the building is colorless to you, since you can't see it, but in the third person subjective frame of mind you know the building is white. (As a side point, the objective status of the building is simply a collection of atoms, all of which must be considered colorless since color only enters once there's perception. This again shows how third person subjective is not objective.)
The base assumptions of relative knowledge are, in general, those things which the imagined person of the third person subjective has assumed. Again this is dependent on context, since the type of third person imagined when speaking to a room of philosophers is much different from the type of third person imagined at a football game. The generic person in each context is assumed to have made very different sorts of assumptions about life, so as a result one's conception of the third person subjective is slightly different in each context. This is why one can make knowledge claims in one situation while not making the claim in the other, and not be inconsistent -- the base of the relative knowledge may be different depending on your conception of the context you're in and the effect that context has in shaping your idea of the third person subjective.
Countering Gettier Counterexamples
Gettier attempts to turn the tripartite analysis of knowledge into a quadripartite analysis. [more info] Once absolute and relative knowledge are understood, however, the analysis of knowledge becomes a bipartite analysis. Knowledge becomes, very simply, justified belief. To include truth as a condition is nonsensical, because deciding if something is true is the same process as deciding if we know something. Truth is the result of the analysis of knowledge, not a part of it... if you have absolute knowledge you've created absolute truth, and if you have relative knowledge that creates relative truth. (This ties in to the Quine-Duhem thesis, which illustrates the way in which truth is relative to auxiliary assumptions.)
Consider a simple Gettier-style counterexample, a person who is said to have a justified true belief that it's 1:15 based on the clock on the wall. The person who is looking at the clock does in fact have relative knowledge that it's 1:15, because they can justify the belief by sense perception of the clock plus the skill learned in kindergarten of how to read clocks, relative to a base which includes that the clock is a functioning non-broken clock, there aren't aliens messing with people's brains, et cetera. In normal cases where the person isn't paranoid these are things which will be accepted as given, due to the fact that people normally expect clocks to work and don't expect to be imposed on by manipulative super-intelligent creatures, and so there's no problem with this sort of base. Note that this assumption that the clock works is obviously a base assumption, and not a part of the actual justification, because the person doesn't investigate for evidence -- if the person did investigate, he or she might interview someone who knows the clock was broken or observe the flaw in the clock directly. In the presented scenario, unsuspicious, the person only wishes to justify knowledge of the current time relative to a base which already includes the assumption of the clock working.
The base assumptions of relative knowledge are, in general, those things which the imagined person of the third person subjective has assumed. Again this is dependent on context, since the type of third person imagined when speaking to a room of philosophers is much different from the type of third person imagined at a football game. The generic person in each context is assumed to have made very different sorts of assumptions about life, so as a result one's conception of the third person subjective is slightly different in each context. This is why one can make knowledge claims in one situation while not making the claim in the other, and not be inconsistent -- the base of the relative knowledge may be different depending on your conception of the context you're in and the effect that context has in shaping your idea of the third person subjective.
It's established that the person knows (has relative knowledge) that it's 1:15. Now, as per Gettier style, we can introduce the "defeater". A Gettier-loyalist, standing outside the room, looks at his own watch and notices it's actually 1:15 just as the person in the room thinks -- but the Gettier-loyalist knows that the clock broke 12 hours ago and therefore it's just a coincidence that the belief "it's 1:15" generated by the clock is true. "Ah ha," says the supporter of Gettier, "this means the person doesn't really know that it's 1:15." This is where an understanding of relative knowledge shows a much more clear picture of what's going on that Gettier's ideas. What we actually have is a situation where the person in the room does not have knowledge relative to the base assumptions of the observing person who knows the clock is broken. This is simply because the observer's base doesn't include the assumption of a functional clock, while the base of the person in the room does include the assumption of a functional clock. Both people agree that the justification process (perception plus clock reading skill) was correct, the disagreement is over whose base is superior. The person inside the room is still fully justified in saying he or she has relative knowledge that it's 1:15, because it is in fact justified relative to his/her base. It's only once the person realizes that there's a problem with their base (perhaps the observer comes in and points out that the clock is broken) that they'll no longer say they had knowledge, because their own base will have changed at that point.
What changes after the defeater is introduced is the conception the person outside the room has of the knowledge of the person inside the room. The way a Gettier supporter would present it would not typically specify an observing person, but this is because they're thinking in the third person subjective -- for the Gettierist, it's either the reader or the third person ghost observer whose knowledge judgment is changed by the defeater, while in the scenario here the judging observer has been materialized into a real person. Relative to the external observer's base, the person whose knowledge the defeater applies to can no longer claim knowledge -- but that's only because for the external observer, those assumptions which are in the base for the person inside the room won't fit in this observer's base and so must be shifted up into the justification. The observer would then likely present the evidence to the person inside the room, and this presentation of evidence would cause the base assumption to be reconsidered and presumably the person inside the room would no longer consider their previous knowledge claim justified relative to their revised base. The result here is that Gettier tells us no more than this fact: relative knowledge is fallible in the sense that it can be contradicted in the future, because a base assumption may not be agreed on as true forever.
Gettier counterexamples simply show that things in your base group of assumptions can later be removed and redefined to be false, or can be considered at the time to other people (who have different bases) as false. The Gettier counterexamples also show that the contents of one person's base, because of evidence they've encountered, may be somewhat different from the contents of another person's base. This is no surprise and really causes no harm at all since the sort of situations detailed by Gettier are unusual, and even in differences that crop up more often we manage to cope most of the time despite our different assumptions. All Gettier illustrates is that relative knowledge is not absolute knowledge... relative knowledge is always fallible. However, unlike with externalism, at least with relative knowledge it's possible to see how and where the fallibility occurs.
Limiting the Use of Externalism
The externalist, although dancing around this issue on all sides, never addresses the most important question of epistemology: "Do I know that P?" When the externalist reliabalist says "I know that P", this has to say something along the lines of "the omniscient ghost which sees these things a priori knows that I am connected to P by a reliable process." Or, as an alternative, "I see through my extrasensory perception and intuition that there's a reliable process connecting me to P." However, when considering that neither the omniscient ghost nor the extrasensory perception ever utters a word, in order to make a claim that something is known a reliablist must say this: "I know through intuiting from extrasensory perception that my process of intuiting from extrasensory perception is a reliable process." Never has there been a more obvious case of begging the question.
One of the complexities of language is that a single word can develop multiple meanings, especially in philosophy. Internalists and externalists seem to have opposite intuitions of what the word "know" means. There are multiple uses of it in common life, which makes it important to distinguish the meanings and important to recognize which applies more often in common speech. A dictionary reveals the common usage of "to know": http://www.dictionary.com/search?q=know . Such a search reveals definitions like these: (1) "To perceive directly; grasp in the mind with clarity or certainty." (2) "To regard as true beyond doubt." (3) To have a practical understanding of, as through experience; be skilled in." 1 and 2 require acquaintance with the justification -- they're about occurent knowledge. 3 is more open to externalist interpretation, but only if it can be described entirely in terms of dispositional knowledge: when put in a certain situation the person would display the behaviors which are associated with seeming to know something. In other words, if it can't be described by a behaviorist it's not something where the dispositional interpretation applies well. The externalist must remain restricted to the relatively rare discussion of dispositional knowledge.
Suppose an externalist brings up the example of an animal's knowledge in an attempt to refute internalism. The externalist points out that we can attribute knowledge to animals, yet animals don't have the mental powers to be aware of the justification of their beliefs. To the externalist, this seems to indicate that the animal needn't be aware of the justification in order to have knowledge and therefore it's the background connection that matters instead of awareness. Of course, the externalist proves to be misguided here once we remember again the nature of relative knowledge and the previously mentioned correlation between intelligence and the size of the group of base assumptions. For a dog, virtually everything has to be in the coherent formation of beliefs which were formed by experience. Deductive powers are very limited for dogs -- the powers exist in a slight sense, but the size of the deductive chains a dog can handle are microscopic compared to what a human can deal with. A dog can has base beliefs about being fed at a certain time every day, and doesn't need to consider these beliefs -- they've formed through repetition and training over the years, in a way a Skinner-type behaviorist could probably get a good start on explaining. There are, however, minor deductive skills on top of this. If a dog sees that the food isn't in the usual spot, it can deduce that it should look around nearby due to the base assumption it has that it's supposed to be fed at that time. (Because it assumes feeding time in the base, the dog is unlikely to be able to understand if you try to explain the various reasons why you were late one particular day. The dog never questions why feeding happens at a certain time, it's not capable of that deduction.) In this way, with most of the process being in the base which was formed for the animal through experience and which the animal isn't capable of questioning and deconstructing in the way humans are capable of questioning and deconstructing their base, it becomes clear how an animal can achieve relative knowledge without a high IQ. The animal can still know that it's time for dinner, with justification perhaps being the way the temperature or location of the sun defines time of day, relative to the base assumption that feeding time must be around that particular time each day. An argument based on animal knowledge, then, does no good for the externalist.
Externalism, as dispositional knowledge, is third person knowledge. As such, it must lie within the self as part of the third person subjective perspective even though it speaks from the external perspective. Externalists and other environmentalist types, taking a behavioristic third person approach, run into behaviorism's problems of defining everything in the third person but not having any self there left to do the defining from. Behaviorism, environmentalism and externalism all define things from the third person perspective, without realizing that the third person is itself only a logical partition of the subjective perspective. This misunderstanding of perspective is what leads the externalist to have no way of describing how the self comes to know something, which in turn leaves no way to describe how the self can make the claim that things are known.
Exernalism has no place in occourent knowledge, since it offers no way to determine when one knows that P without begging the question, and without occourent knowledge dispositional knowledge could not even exist. Dispositional knowledge lies in the third person subjective, so it requires an observing person (or in abstract cases, an abstract and ghost like ideal observer person) to be aware of it. Thus, our only place to use externalism is as a way of speaking dispositionally within internalism.
Externalism is motivated by the desire to escape the global skeptic (although doing this by losing the ability to speak of how the self knows something is like escaping the ravenous skeptic by hiding in his stomach), and due to this externalists will often tell internalists "Well, if you don't like externalism, how are you going to avoid the global skeptic? Silence? Thought as much, you don't have a better theory so stop criticizing." With relative knowledge, however, the global skeptic simply becomes someone who's not very useful to talk to since you can't convince him/her of anything and he or she has nothing useful to add to anything. There's nothing about the skeptic which needs refuting, they're simply base-impaired people who serve no function in normal life but may keep in check any philosopher who begins to go off the deep end by claiming absolute knowledge. To relative knowledge, the global skeptic can do nothing... if the skeptic wishes to get involved in the discussion he must for the sake of the argument accept the base assumptions -- otherwise, unless he has empirical evidence which will cause people to want to revise their base, the skeptic might as well be speaking another language and can be ignored. Thus, an internalist who understands the distinction between relative and absolute knowledge needn't worry about the global skeptic.
What's been created here is a way of doing epistemology that merges foundationalism with coherentism. It can be classified as a falliblist foundationalist system that incorporates a form of coherentism as the way to determine which beliefs are basic. It keeps the deductively useful benefits of foundationalism, solves the problem of the properly basic belief and simplifies analysis of knowledge. Most importantly, it aims at achieving an understanding of what's behind the words when a common person makes a statement about knowledge. Interpreting the claim "I know that P" is what any theory of epistemology needs to accomplish, and an understanding of relative knowledge makes sense out what a person means when claiming "I know that P." It offers a way of looking at common talk -- not only the ideal situations where people agree, but also the situations where people cannot agree because they have different base assumptions, even if they agree about the logic of the justification process.
