A Critique of Ayn Rand's Contextual Theory of Knowledge

[George H. Smith]

This is another excerpt (without formatting and endnotes) from my book Why Atheism? And again, I am posting it because of its relevance to a discussion in the Moral Certainty thread. Keep in mind that this is merely the last section of a chapter that deals with the nature of knowledge.

From Chapter Four, "Belief and Knowledge."

VI

Ayn Rand’s Theory of Knowledge

Ayn Rand’s theory of truth challenges the conventional theory of knowledge that I am here defending. Rand does not agree that knowledge must be both justified and true; knowledge is merely justified belief and nothing more. Why? Because we have no way of ascertaining what is true except through justification, so to say that a belief, if it is to qualify as knowledge, must also be “true,” is technically redundant. For truth, like justification, is contextual; it depends on the evidence that is available to us at a given time.

This position merits examination for two reasons: first, because it an important element of Rand’s contextualism, a theory of knowledge that is plausible and rich in implications; and, second, because it reveals a latent tendency towards epistemological relativism that sharply conflicts with Rand’s spirited insistence that truth is “objective.” (The philosopher Leonard Peikoff, a confidant of Rand’s for many years, has extended Rand’s contextualism beyond that which appears in Rand’s published writings, so I shall rely on his treatment.)

Truth is defined by Ayn Rand as “the recognition of reality” – and this definition, according to Peikoff, is “in essence…the traditional correspondence theory of truth.” This is true to a point, but there is an interesting aspect to Rand’s theory that distinguishes from the approach I am defending here.

Rand, as we have seen, disagrees with the view that knowledge is true and justified belief. Her reason for treating truth and justification as virtually synonymous is a compelling one, namely. that we cannot know what is true except in the context of knowledge that is available to us at any given time. All judgments of truth by which we justify a belief are contextual, so to treat truth as if it were an abstract correspondence between a proposition and a fact, divorced from the particular judgments of concrete individuals, is implicitly to establish omniscience as a standard of human knowledge. As Peikoff puts it:

“There can be no ‘correspondence’ or ‘recognition’ without the mind that corresponds or recognizes….The true is identified by reference to a body of evidence; it is pronounced ‘true’ because it can be integrated without contradiction into a total context.”

This is the foundation for Rand’s contextual theory of certainty. Human knowledge is necessarily limited, which means that man has “a specific cognitive context” at every stage in the development of his knowledge. If, therefore, an idea can be traced to its foundation in sense data and is based on sufficient evidence, then that idea has been “validated.” Again quoting Peikoff, “Logical processing of an idea within a specific context of knowledge is necessary and sufficient to establish the idea’s truth.”

This statement, through reasonable on its face, leads to the rather peculiar conclusion – peculiar at least for those who stress the objectivity of knowledge – that there exist different truths for people who work from different contexts of knowledge. Consider one of Peikoff’s examples: the belief of early medical researchers that four types of blood (A, B, AB, and O), while incompatible with each other, are each compatible with their own type. It was later discovered that this was not always the case: a recipient of blood from a donor with the same type occasionally responded negatively – a problem that was later explained by the RH factor, which is present in the blood of some individuals but not others.

The philosophical question raised by this story is this: Was the early belief -- that each blood type is compatible with its own type – true or false? According to the conventional view (which I am defending), this belief, though justified given the information available to researchers at the time, was in fact false, because it did not take into account the RH factor, which was discovered later. But Peikoff disagrees. Given the knowledge available to the earlier researchers, their belief that type A bloods are compatible was justified. Thus, within that context, the proposition, “A bloods are compatible” was true. As Peikoff says:

“This proposition represented real knowledge when it was first reached, and it still does so; in fact, like all properly formulated truths, this truth is immutable. Within the context initially specified, A bloods are and always will be compatible.” (My emphasis.)

This is a difficult passage to interpret reasonably, since it seems so obviously wrong. If “within the context initially specified, A bloods are and always will be compatible,” then what was the original problem that caused researchers eventually to discover the RH factor, if not the fact that A bloods were sometimes incompatible? The perception of a problem -- that a blood type was not always compatible with the same type -- necessarily preceded the search for a solution. Researchers had to become convinced that their current theory was false (at least in some respects) or they would never have looked for a better one. If Peikoff is correct, if the initial theory was immutably true within the context of knowledge available to medical researchers at that time, then there would have been no reason (and no motive) for them to improve upon that theory.

If, contrary to Peikoff (and presumably to Rand), we distinguish between truth and justification, then we can explain the preceding example without difficulty. The initial belief about blood types was justified, given the information available to early researchers, but it was not true, as researchers themselves later discovered.

Why is this commonsensical analysis objectionable to the Randian contextualist? Because it allegedly divorces the notion of truth from the knowing mind, exiling it to an ethereal world of Platonic essences. But since there is no “truth” apart from the particular recognition of a fact of reality, and since this presupposes a rational being with a specific mode of cognition, our concept of truth, if it is to have meaning, must be grounded in the fallible and limited nature of human reason. Thus if we define “truth” as the abstract correspondence between a proposition and a fact, while neglecting the cognitive process of recognition by which facts are identified, we transform truth into an unattainable ideal that will forever lie beyond our grasp.

This is a reasonable concern. Most forms of epistemological skepticism are based on the fallibility and limitations of human reason. It has been claimed, for example, that we can never be certain of anything, because many beliefs from earlier ages, which were accepted with the highest degree of certitude, were subsequently proven to be false. And given our innate fallibility, our present beliefs may also be proven wrong by subsequent generations, which means that we should not claim certainty for any of them.

Though it is commendable that Rand and Peikoff wish to extirpate this kind of facile skepticism by pointing to its illicit reliance on infallibility as a criterion of certainty (a point that has been made by many other philosophers), this by no means requires that we reduce truth to sufficient justification, while dispensing altogether with the traditional notion of abstract correspondence. While we should not permit fallibility to be used as a pretext for skepticism, we should also avoid the opposite error of proclaiming as an “immutable truth” every justified belief that is subsequently revised or rejected.

There is a sense in which it redundant to define knowledge as “justified and true belief,” because there is no royal road to truth apart from the justification we have to affirm the truth of a particular proposition. We cannot somehow circumvent our best available evidence in favor of proposition and ascertain its truth directly, without the mediation of a cognitive process.

This applies even to self-evident truths, such as the Law of Identity (A is A) and other logical axioms. The traditional philosophical distinction between intuitive reason (which is able to grasp self-evident truth immediately) and discursive reason (which progresses in a chain of reasoning from premises to conclusion) is highly misleading at best. The axioms of logic, though first in the order of knowledge, are not first in the order of time. That is to say, although these axioms are logically presupposed by all other knowledge (in the sense that no valid knowledge can contradict them), they are not the first knowledge we acquire as infants. By the time we study logic (if we ever do), we already have a store of knowledge that we bring to bear in assessing its axioms, and in light of which we pronounce them “self-evident” – a term that would be meaningless if not set in opposition to truths that are not self-evident. Thus the axioms of logic, though self-evidently true, cannot be identified as such without a cognitive process. It is in this sense that all reasoning is discursive, and none is intuitive. No knowledge is given to man automatically, without mental labor.

What, then, is the point of retaining abstract truth in our definition of knowledge, if we have cannot ascertain what is true apart from our contextually justified beliefs? The point is simply this: truth and justified belief are not the same thing, so they need to be distinguished. Many justified beliefs – propositions that were perfectly reasonable to believe in a particular context – have been not only revised (as Peikoff seems to think) but also rejected completely.

The Copernican revolution in astronomy, which was followed by Kepler’s rejection of circular orbits, is a clear example of this. Do we seriously wish to say that the medieval cosmology – an eclectic brew of Aristotelian physics, Neoplatonic metaphysics, and Ptolemaic astronomy -- was (and is) “true,” given the medieval context of knowledge and the best evidence that was then available? Do we seriously wish to say that the doctrine of circular planetary orbits was (and is) “true,” given the fact it was supported by a long-established and coherent theory of metaphysics, was defended for centuries by the best scientific minds, and was quite adequate for explaining the best available astronomical data (prior to the more precise observations of Tycho Brahe)?

We might say that these and other scientific theories – which have not been merely revised, but completely discarded – were justified beliefs for medieval thinkers, given the apparent evidence in their favor and their overall coherence with the medieval worldview. But they were false nonetheless, absolutely and unequivocally, however justified they may have been at one time. They were not somehow “contextually” true, much less “immutably” so.

Of course, a ready reply to this argument is available to the defenders of Randian contextualism. It can be said (and often has been said) that the medieval cosmology was sustained by a rigid ecclesiastic orthodoxy during an age in which there was little if any authentic science. In this view, the scientific revolution was the triumph of honest inquiry, empiricism. and experimentation over the rigid dogmas of religious orthodoxy. Thus when confronted with the historical transition to modern science, the Randian contextualist might argue that the medieval cosmology does not qualify as even contextually true, because it was not based on logical cognition and authentic evidence. Modern science, therefore, was not a revision of medieval science, because the latter was not legitimate science at all.

This or any similar reply will not solve the incipient relativism of Randian contextualism, even if we accept the preceding characterization of the medieval cosmology at face value (which we should not). The Randian contextualist cannot pick and choose his contextual and immutable truths, depending on whether they approximate modern beliefs, for this would unfairly subject medieval thinkers to the same standards of infallibility and omniscience against which the Randians so vigorously (and rightfully) protest.

A contextual theory of knowledge, in my judgment, must strike a delicate balance between relativism and absolutism. And this is precisely why we should retain the traditional view that knowledge is justified and true belief. Justification is relative, whereas truth is absolute. That is to say, what counts as adequate justification for a belief may be relative to the available evidence and one’s context of knowledge, whereas the truth of a belief is absolute: a proposition either corresponds to a fact or it does not, and this matter has nothing to do with the relative justification for a belief.

It is by blending the relativity of justification with the absolutism of truth that we arrive at a true contextual theory of knowledge. Justification without truth leads to a futile relativism, while truth without justification, by equating knowledge with infallibility, leads to a skepticism that is equally futile.

Although we have no royal roads to knowledge, although we cannot know what is true apart from what we are justified in believing to be true, this does not mean that we can, or should, dispense with the notion of absolute truth. This notion, if it is an abstract ideal, also functions as a concrete reminder of our fallibility. It stands, like Mordicai at the gate, as a reminder that no belief can claim a privileged immunity from critical evaluation; and that every reflective person, however justified his beliefs may be, is prey to the same errors of fallibility as everyone else.

[tjohnson]

It would be best to abandon the notion of 'truth' altogether and use the notion of 'structural similarity' instead. All these "problems" would then vanish. Knowledge is like a map and as time goes on our maps get better and better - more accurate. They will never be completely accurate, which would correspond to "true" in the old way of thinking, but they will continue to get better.

[BaalChatzaf]

Although we have no royal roads to knowledge, although we cannot know what is true apart from what we are justified in believing to be true, this does not mean that we can, or should, dispense with the notion of absolute truth. This notion, if it is an abstract ideal, also functions as a concrete reminder of our fallibility. It stands, like Mordicai at the gate, as a reminder that no belief can claim a privileged immunity from critical evaluation; and that every reflective person, however justified his beliefs may be, is prey to the same errors of fallibility as everyone else.

Word! Nachon v'emmet!

Ba'al Chatzaf

[Brant Gaede]

It would be best to abandon the notion of 'truth' altogether and use the notion of 'structural similarity' instead. All these "problems" would then vanish. Knowledge is like a map and as time goes on our maps get better and better - more accurate. They will never be completely accurate, which would correspond to "true" in the old way of thinking, but they will continue to get better.

This is true.

Opps! My bad.

--Brant

[Brant Gaede]

Truth is not "contextual." Truth is reality correspondence. It is knowledge that is contextual, and tentative. We (should) seek truth. What we get is knowledge, a lot of it by bumping into things. The 20th C. generally revealed the truth about totalitarianism and the moral bankruptcy of recreating and molding human beings and society from the top down. On the psychological level, Rand comes up short positing human perfectibility as an ideal. The denizens of Galt's Gulch are hardly more differentiated from each other than the collectivist citizens of Anthem. The saving grace is she was a champion of freedom and in freedom people can make themselves and be themselves NIOF.

--Brant

[tjohnson]

Truth is not "contextual." Truth is reality correspondence. It is knowledge that is contextual, and tentative. We (should) seek truth. What we get is knowledge, a lot of it by bumping into things. The 20th C. generally revealed the truth about totalitarianism and the moral bankruptcy of recreating and molding human beings and society from the top down. On the psychological level, Rand comes up short positing human perfectibility as an ideal. The denizens of Galt's Gulch are hardly more differentiated from each other than the collectivist citizens of Anthem. The saving grace is she was a champion of freedom and in freedom people can make themselves and be themselves NIOF.

--Brant

Brant, your elucidation is sometimes breathtaking !

[tjohnson]

It would be best to abandon the notion of 'truth' altogether and use the notion of 'structural similarity' instead. All these "problems" would then vanish. Knowledge is like a map and as time goes on our maps get better and better - more accurate. They will never be completely accurate, which would correspond to "true" in the old way of thinking, but they will continue to get better.

This is true.

Opps! My bad.

--Brant

This is similar in structure.

[George H. Smith]

It would be best to abandon the notion of 'truth' altogether and use the notion of 'structural similarity' instead. All these "problems" would then vanish.

Yeah, right. Substituting two words for one would solve everything.

Knowledge is like a map and as time goes on our maps get better and better - more accurate.

Metaphors won't get us very far in this area. We could also say that knowledge is like a compass, or like a beam of light that guides us through darkness, etc., etc.

Moreover, if we can compare the "structural similarities" (whatever that is supposed to mean) between our "map" and reality, this presupposes that we can consult reality directly, in which case we don't need a "map" in the first place.

We stand to real maps as outside observers -- something that is impossible with knowledge claims. Moreover, real maps, if they are to be accurate, presuppose knowledge of reality. Without this prior knowledge, we could never assess the accuracy of a given map. So how do we gain knowledge of our metaphorical map? With yet another metaphorical map?

Problems like this are legion when metaphors are substituted for careful thinking about the nature of knowledge. But if we must use a metaphor, the metaphor of "light" (which was very popular during the Enlightenment) is preferable to a "map." A beam of light enables us to see the world (part of it, anyway), but we don't "see" reality through a map. On the contrary, we must first be able to see before we can "see" the supposed "structural similarities" between a metaphorical map and reality.

I'm frankly not even sure why I'm responding to the claim that a map metaphor would cause "all" the problems of epistemology to "vanish." The claim is absurd on its face.

Ghs

[Daniel Barnes]

Ayn Rand’s Theory of Knowledge

I think I agree with most of this critique. Which is very boring of me. However I will not be able to discuss for a couple of days due to other commitments. So my misc. caviling will have to wait.

[tjohnson]

Moreover, if we can compare the "structural similarities" (whatever that is supposed to mean) between our "map" and reality, this presupposes that we can consult reality directly, in which case we don't need a "map" in the first place.

Well, suppose you have a map that says Boston is south of New York and you are in New York and you head south to get to Boston and you never get there. We check the validity of the map (or theory) by trying it out. Not sure what you mean about consulting "reality directly"??

[Donovan A.]

Hello there,

I'd like to address a few sections of your post, if I may:

Ayn Rand’s theory of truth challenges the conventional theory of knowledge that I am here defending. Rand does not agree that knowledge must be both justified and true; knowledge is merely justified belief and nothing more. Why? Because we have no way of ascertaining what is true except through justification, so to say that a belief, if it is to qualify as knowledge, must also be “true,” is technically redundant. For truth, like justification, is contextual; it depends on the evidence that is available to us at a given time.

The following I propose as simple referenced definitions of terminology:

"Knowledge" is ... a mental grasp of a fact(s) of reality, reached either by perceptual observation or by a process of reason based on perceptual observation. AR, "Concepts of Consciousness," ITOE

Implicit knowledge is passively held material which, to be grasped, requires a special focus and process of consciousness. AR. "Axiomatic Concepts,"ITOE,

Truth is the recognition of reality .

AIt GS. FNI. 126. LP discusses in "The Arbitrary as Neither True Nor False ," OPAR.

This position merits examination for two reasons: first, because it an important element of Rand’s contextualism, a theory of knowledge that is plausible and rich in implications; and, second, because it reveals a latent tendency towards epistemological relativism that sharply conflicts with Rand’s spirited insistence that truth is “objective.” (The philosopher Leonard Peikoff, a confidant of Rand’s for many years, has extended Rand’s contextualism beyond that which appears in Rand’s published writings, so I shall rely on his treatment.)

I would like to note that the concept of "objective" (as I understand it) in this context means that reality is independent from consciousness. A statement or conclusion is true only if it adheres to reality as opposed to consciousness or whim, wish or belief. Truth is derived from the concept of existence. Falseness, is derived from the concept of non-existence.

This statement, through reasonable on its face, leads to the rather peculiar conclusion – peculiar at least for those who stress the objectivity of knowledge – that there exist different truths for people who work from different contexts of knowledge. Consider one of Peikoff’s examples: the belief of early medical researchers that four types of blood (A, B, AB, and O), while incompatible with each other, are each compatible with their own type. It was later discovered that this was not always the case: a recipient of blood from a donor with the same type occasionally responded negatively – a problem that was later explained by the RH factor, which is present in the blood of some individuals but not others.

The philosophical question raised by this story is this: Was the early belief -- that each blood type is compatible with its own type – true or false? According to the conventional view (which I am defending), this belief, though justified given the information available to researchers at the time, was in fact false, because it did not take into account the RH factor, which was discovered later. But Peikoff disagrees. Given the knowledge available to the earlier researchers, their belief that type A bloods are compatible was justified. Thus, within that context, the proposition, “A bloods are compatible” was true. As Peikoff says:

“This proposition represented real knowledge when it was first reached, and it still does so; in fact, like all properly formulated truths, this truth is immutable. Within the context initially specified, A bloods are and always will be compatible.” (My emphasis.)

This is a difficult passage to interpret reasonably, since it seems so obviously wrong. If “within the context initially specified, A bloods are and always will be compatible,” then what was the original problem that caused researchers eventually to discover the RH factor, if not the fact that A bloods were sometimes incompatible? The perception of a problem -- that a blood type was not always compatible with the same type -- necessarily preceded the search for a solution. Researchers had to become convinced that their current theory was false (at least in some respects) or they would never have looked for a better one. If Peikoff is correct, if the initial theory was immutably true within the context of knowledge available to medical researchers at that time, then there would have been no reason (and no motive) for them to improve upon that theory.

In response to this, I'd simply like to post more from that same passage:

The principle here is evident: since a later discovery rests hierarchically on earlier

knowledge, it cannot contradict its own base. The qualified formulation in no way clashes with

the initial proposition, viz.: "Within the context of the circumstances so far known, A bloods

are compatible." This proposition represented real knowledge when it was first reached, and it

still does so; in fact, like all properly formulated truths, this truth is immutable. Within the

context initially specified, A bloods are and always will be compatible.

The appearance of a contradiction between new knowledge and old derives from a single

source: context-dropping. If the researchers had decided to view their initial discovery as an

out-of-context absolute; if they were to declare—in effect, as a matter of dogma:

"A bloods will always be compatible, regardless of altered circumstances"; then of course the

next factor discovered would plunge them into contradiction, and they would end up

complaining that knowledge is impossible. But if a man reaches conclusions logically and

grasps their contextual nature, intellectual progress poses no threat to him; it consists to a great

extent in his identifying ever more fully the relationships, the connections among facts, that

make the world a unity. Such a man is not dismayed to find that he always has more to learn.

He is happy about it, because he recognizes that he is expanding and refining his knowledge,

not subverting it.

Although the researchers cannot claim their discovery as an out-of-context absolute, they

must treat it as a contextual absolute (i.e., as an immutable truth within the specified context).

The researchers must know that the initial generalization is valid—"know" as against

guess, hope, or feel. It is only on this basis that they can progress to further discoveries. Since

it is an established truth that A bloods are compatible under the circumstances so far

encountered, the researchers are able to infer, when they observe a new reaction, the presence

of a new factor. By contrast, when the anti-contextual mentality observes the new reaction, he

stops dead. "My generalization was unreliable," he sighs, "science is a progression of exploded

theories, everything is relative." - OPAR p.174

If, contrary to Peikoff (and presumably to Rand), we distinguish between truth and justification, then we can explain the preceding example without difficulty. The initial belief about blood types was justified, given the information available to early researchers, but it was not true, as researchers themselves later discovered.

By justification I believe you mean evidence. A conclusion without evidence is a floating abstraction, a dogma. One has no bases for accepting a conclusion divorced from facts, i.e., sensory evidence or logical demonstration.

This is a reasonable concern. Most forms of epistemological skepticism are based on the fallibility and limitations of human reason. It has been claimed, for example, that we can never be certain of anything, because many beliefs from earlier ages, which were accepted with the highest degree of certitude, were subsequently proven to be false. And given our innate fallibility, our present beliefs may also be proven wrong by subsequent generations, which means that we should not claim certainty for any of them.

See The Fallacy of the Stolen Concept by Nathaniel Branden.

A contextual theory of knowledge, in my judgment, must strike a delicate balance between relativism and absolutism. And this is precisely why we should retain the traditional view that knowledge is justified and true belief. Justification is relative, whereas truth is absolute. That is to say, what counts as adequate justification for a belief may be relative to the available evidence and one’s context of knowledge, whereas the truth of a belief is absolute: a proposition either corresponds to a fact or it does not, and this matter has nothing to do with the relative justification for a belief.

Many people in our Kantian era think, mistakenly, that absolutism is

incompatible with a contextual approach to knowledge. These people define an "absolute" as a

principle independent of any other fact or cognition; i.e., as something unaffected by anything

else in reality or in human knowledge. Such a principle could come to be known only by

revelation. An eloquent example of this approach was offered years ago by a famous relativist,

who told his class that airplanes refute the law of gravitation. Gravitation, he explained, means

that entities over a certain weight fall to the earth; but an airplane in flight does not. Someone

objected that there are many interacting factors in reality, and that gravitation involves an

object's falling only if the gravitational pull is not counteracted by an opposing force, as it is in

the airplane's case. To which the professor replied: "Precisely. Gravitation is conditional; its

operation depends on circumstances; so it is not an absolute." What then would qualify as an

absolute? Only a fact that has no relationships to anything (like Hegel's supernatural Absolute).

Such a fact would be knowable only "in itself," by mystic insight, without the "contamination"

of any "external" context of evidence.

The modern definition of "absolute" represents the rejection of a rational metaphysics and

epistemology. It is the inversion of a crucial truth: relationships are not the enemy of

absolutism; they are what make it possible. We prove a conclusion on the basis of facts

logically related to it and then integrate it into the sum of our knowledge. That process is what

enables us to say: "Everything points to this conclusion; the total context demands it; within

these conditions, it is unshakable." About an isolated revelation, by contrast, we could never be

secure. Since we would know nothing that makes it so, we could count on nothing to keep it so,

either.

Contextualism does not mean relativism. It means the opposite. The fact of context does

not weaken human conclusions or make them vulnerable to overthrow. On the contrary,

context is precisely what makes a (properly specified) conclusion invulnerable.

So far, I have considered only two mental states, knowledge and ignorance, and two

corresponding verdicts to define an idea's status: "validated" or "unknown." Inherent in the

mind's need of logic, however, is a third, intermediate status, which applies for a while to

certain complex higher-level conclusions. In these cases, the validation of an idea

is gradual; one accumulates evidence step by step, moving from ignorance to knowledge

through a continuum of transitional states. The main divisions of this continuum (including its

terminus) are identified by three concepts: "possible," "probable," and "certain."

The first range of the evidential continuum is covered by the concept "possible." A

conclusion is "possible" if there is some, but not much, evidence in favor of it, and nothing

known that contradicts it. This last condition is obviously required—a conclusion that

contradicts known facts is false—but it is not sufficient to support a verdict of "possible."

There are countless gratuitous claims in regard to which one cannot cite any contradictory fact,

because they are inherently detached from facts; this does not confer on such claims any

cognitive status. For an idea to qualify as "possible," there must be a certain amount of

evidence that actually supports it. If there is no such evidence, the idea falls under a different

concept: not "possible," but "arbitrary." - OPAR p.174-176

Edited March 1, 2010 by Randall

[rhartford]

I'm frankly not even sure why I'm responding to the claim that a map metaphor would cause "all" the problems of epistemology to "vanish." The claim is absurd on its face.

If you are like me, you answer because you are patient and you know that some others will find your comments valuable.

Your analysis of "contextual certainty" to start this thread was quite interesting. Your statement that "although we cannot know what is true apart from what we are justified in believing to be true, this does not mean that we can, or should, dispense with the notion of absolute truth. This notion, if it is an abstract ideal, also functions as a concrete reminder of our fallibility" sounds right. The thought that truth is possible also spurs us to greater and more careful effort to use cognitive processes, objective processes, that best reach truth.

Truth serves a similar role epistemologically that existence serves metaphysically. Certainly we can be certain of something. In fact I know something about everything. Everything that is exists, and I am certain of the truth of that statement.

[George H. Smith]

Moreover, if we can compare the "structural similarities" (whatever that is supposed to mean) between our "map" and reality, this presupposes that we can consult reality directly, in which case we don't need a "map" in the first place.

Well, suppose you have a map that says Boston is south of New York and you are in New York and you head south to get to Boston and you never get there. We check the validity of the map (or theory) by trying it out. Not sure what you mean about consulting "reality directly"??

Suppose you try out a real map and find it reliable. The map is reliable only because someone -- i.e., the mapmaker -- used his knowledge to render the map reliable. And the knowledge of the (original) mapmaker did not come from another map. If the original mapmaker could not attain the required knowledge (distances, etc.) without the aid of another map, he would be unable to make a map in the first place. Maps are merely aids that presuppose that knowledge can be acquired by other means.

Moreover, suppose I use a map to get to an address across town. At the completion of my trip, I do not verify that I have reached the correct address by consulting my map; rather, I consult reality directly by looking at the street address.

Lastly -- and as I indicated in my previous post -- we consult maps as outside observers. If we want to determine the accuracy of a map, we look at a map and we look at reality, and we then decide whether or not the two correspond.

But this external, objective viewpoint is not possible with the metaphorical "map" of knowledge. If all we have are maps, and if we can never see outside these maps (i.e, acquire knowledge without their aid), then we could never compare those maps to reality to determine if they correspond to one another. This would be like using a real map to drive to a destination while never looking at anything except the map along the way. We never look at reality (roads, street signs, etc.); all we do is stare at the map. This would be like driving with a blindfold on. We would barely get out of the driveway (for which we don't need a map) before crashing into something.

Ghs

[tjohnson]

Suppose you try out a real map and find it reliable. The map is reliable only because someone -- i.e., the mapmaker -- used his knowledge to render the map reliable. And the knowledge of the (original) mapmaker did not come from another map. If the original mapmaker could not attain the required knowledge (distances, etc.) without the aid of another map, he would be unable to make a map in the first place. Maps are merely aids that presuppose that knowledge can be acquired by other means.

Really? If I was making a map the first thing I would do is look at existing ones and see where mine is similar and where mine is different. An explorer would take an existing map and while using it he would see where it needed work and he would be in a good position to make corrections since he was currently "in the territory". So yes, the first mapmaker require investigating the territory and so do all subsequent revisions.

[tjohnson]

But this external, objective viewpoint is not possible with the metaphorical "map" of knowledge.

I don't understand this statement. We have a "map" by Newton concerning the 'the addition of velocities' and we have one by Einstein which differs in structure (includes factors which Newton did not consider). We can "refer to the territory" (perform experiments) and conclude that Einstein's is in fact more accurate.

[tjohnson]

The thought that truth is possible also spurs us to greater and more careful effort to use cognitive processes, objective processes, that best reach truth.

Why couldn't the formulation of 'structural similarity' serve this purpose just as well as 'truth'?

[BaalChatzaf]

The thought that truth is possible also spurs us to greater and more careful effort to use cognitive processes, objective processes, that best reach truth.

Why couldn't the formulation of 'structural similarity' serve this purpose just as well as 'truth'?

Structural similarity is the nub of the Correspondence Theory of Truth.

Ba'al Chatzaf

[tjohnson]

The thought that truth is possible also spurs us to greater and more careful effort to use cognitive processes, objective processes, that best reach truth.

Why couldn't the formulation of 'structural similarity' serve this purpose just as well as 'truth'?

Structural similarity is the nub of the Correspondence Theory of Truth.

Ba'al Chatzaf

Yes, but without that problematic term 'truth'. We have seen that if a statement is deemed TRUE at one point in time the very same statement may be deemed FALSE at another point in time. 'Structural similarity', on the other hand allows for revision over time.

[George H. Smith]

Suppose you try out a real map and find it reliable. The map is reliable only because someone -- i.e., the mapmaker -- used his knowledge to render the map reliable. And the knowledge of the (original) mapmaker did not come from another map. If the original mapmaker could not attain the required knowledge (distances, etc.) without the aid of another map, he would be unable to make a map in the first place. Maps are merely aids that presuppose that knowledge can be acquired by other means.

Really? If I was making a map the first thing I would do is look at existing ones and see where mine is similar and where mine is different. An explorer would take an existing map and while using it he would see where it needed work and he would be in a good position to make corrections since he was currently "in the territory". So yes, the first mapmaker require investigating the territory and so do all subsequent revisions.

You don't have a clue why examples like this generate problems for your map metaphor, do you?

From now on, I think I'll use the metaphor of a can-opener to show how we can make all epistemological problems "vanish." We want food, just as we want knowledge. We use an opener to open a can and get to the food inside. Similarly, if we want to acquire knowledge, all we need do is open the metaphorical can of the unknown with the metaphorical can-opener of reason and see if there is knowledge to be found inside.

Or maybe the metaphor of a microwave oven would work better. We need to process our sense data to make it useful. Similarly, we need to process a frozen dinner to make it edible. So we pop the frozen dinner into a microwave oven, and presto! -- problem solved.

Now all I need is a catchy saying, such as "The microwave oven is not the frozen dinner." Yeah, that's the ticket!

I like this metaphor thing. It sure beats exerting the mental labor needed for a careful and sustained analysis of epistemological problems.

Ghs

[merjet]

Structural similarity is the nub of the Correspondence Theory of Truth.

Huh? http://en.wikipedia.org/wiki/Structural_similarity

Structural similarity is the nub of the Correspondence Theory of Truth.

Yes, but without that problematic term 'truth'. We have seen that if a statement is deemed TRUE at one point in time the very same statement may be deemed FALSE at another point in time. 'Structural similarity', on the other hand allows for revision over time.

Where did you get the notion that the correspondence theory of truth isn't time-dependent? The classic statement of that theory is by Aristotle. "To say that that which is is not or that which is not is, is a falsehood; and to say that that which is is and that which is not is not, is true" (Metaphysics, 1011b26).

Aristotle said "is", not "is and always is". Obviously Aristotle would not have said "Plato is sitting" is always true.

[George H. Smith]

But this external, objective viewpoint is not possible with the metaphorical "map" of knowledge.

I don't understand this statement. We have a "map" by Newton concerning the 'the addition of velocities' and we have one by Einstein which differs in structure (includes factors which Newton did not consider). We can "refer to the territory" (perform experiments) and conclude that Einstein's is in fact more accurate.

All you have done in your explanation is to tack on some needless verbiage about a metaphorical map and "structural similarities." Everything that needs to be explained in your example can be explained by conventional epistemological theory, such as the correspondence theory of truth.

Ghs

[tjohnson]

Aristotle said "is", not "is and always is". Obviously Aristotle would not have said "Plato is sitting" is always true.

Hardly a good example of 'knowledge' but what the heck If Aristotle didn't mean "Plato is and always is sitting" then he should have said "Plato is sitting now", or even better "it appears to me that Plato is sitting now".

[merjet]

Aristotle said "is", not "is and always is". Obviously Aristotle would not have said "Plato is sitting" is always true.

If Aristotle didn't mean "Plato is and always is sitting" then he should have said "Plato is sitting now", or even better "it appears to me that Plato is sitting now".

Adding "now" is redundant; "is" is present tense. Adding "appears to me" implies some doubt; to include it when there is no doubt is silly.

[Robert Campbell]

This or any similar reply will not solve the incipient relativism of Randian contextualism, even if we accept the preceding characterization of the medieval cosmology at face value (which we should not). The Randian contextualist cannot pick and choose his contextual and immutable truths, depending on whether they approximate modern beliefs, for this would unfairly subject medieval thinkers to the same standards of infallibility and omniscience against which the Randians so vigorously (and rightfully) protest.

George,

Thanks for posting this. It is concise and timely, and provides a good starting point for discussion.

From my historical research on the doctrine of the arbitrary assertion, I'm inclined to think that contextual certainty is similar to the latter-day, more extreme view of arbitrariness—both are Leonard Peikoff's creations. Even though Rand endorsed them, they're more accurately called Peikovian than Randian.

The Peikovian doctrine about certainty causes trouble in two directions.

Applied retrospectively, it leads, as you've noted, to relativism (i.e., theories that have been overthrown, or undergone radical surgery, were true in their epistemological context, but not in ours).

Or it leads to retroactive downgrading of the theory that's been overthrown (it was adopted irrationally, no one knew how to be scientific yet, etc.).

I've run into the retroactive downgrading in a number of online discussions. A common reaction to the phlogiston theory of combustion among self-declared Objectivists is that the theory turned out to be wrong because it wasn't scientific and the phlogiston theorists were irrational. Since the phlogiston theory was developed in the early 1700s, after the advent of modern physics and astronomy, and its promotion and adoption had no obvious religious implications, they really have to stretch to make such claims.

I recall an exchange with the late Ron Merrill, who endorsed the pre-scientificity of phlogiston theory. Merrill ended up branding pre-Einsteinian physics inadequately rational and scientific. I think he'd gone beyond stretching, and had traveled some distance into Untenable Land.

The same style of reasoning would require the downgrading of superseded points of view (to unscientific or irrational) whenever a major change in theory takes place in the future. So conceptions that are supposedly contextually certain today may eventually be rejected, not merely as erroneous, but as ill-thought-out and misguided.

Robert Campbell

PS. There's a further complication for many in the ARI crowd, because they subscribe to another Peikovian doctrine: 20th century scientific decadence. I expect to see the physics post-1905 run down pretty sharply in David Harriman's forthcoming book.

[tjohnson]

All you have done in your explanation is to tack on some needless verbiage about a metaphorical map and "structural similarities." Everything that needs to be explained in your example can be explained by conventional epistemological theory, such as the correspondence theory of truth.

In conventional epistemological theory is there a place for partially true? One could say Newton's equations for addition of velocity are partially true because they work when the speeds involved are relatively small.

Edited March 1, 2010 by general semanticist