Objectivism and the Correspondence Theory of Truth

[dan_edge]

By Dan Edge from The Edge of Reason,cross-posted by MetaBlog

Introduction

This essay is a discussion of the Correspondence Theory of Truth (CTOT) and how the theory fits into the philosophy of Objectivism. I will argue that properly understood, the CTOT is compatible with Objectivist epistemology. I will also discuss pitfalls that must be avoided to keep the door closed to skepticism.

Readers should note that I am writing here for two distinct audiences. The first is the Objectivist community at large, those who are already familiar with the philosophy. I am also presenting this essay to my Senior Seminar in Philosophy class at the University of South Carolina for peer review.

To the Objectivist community: Some have expressed dissatisfaction with an unqualified acceptance of the Correspondence Theory, and with good reason, as I hope to demonstrate. A proper understanding of the Correspondence Theory is necessary to defend against Juggernaut of skepticism that has been sweeping the academic community for the past century.

To my classmates: This paper is based on information contained in my in-class presentation of Objectivist Epistemology, and it assumes the truth of Objectivism. I will glaze over some technical aspects of Objectivist principles. I encourage any of you to approach me with questions about these principles as you are working through the text.

----------------

The Correspondence Theory of Truth

The first formal expression of the Correspondence Theory of Truth (CTOT) can be traced back to Aristotle, who wrote: “To say of what is that it is not, or of what is not that it is, is false, while to say of what is that it is, and of what is not that it is not, is true” (Metaphysics 1011b25), though Plato wrote very similar formulations (Cratylus 385b2, Sophist 263b). Kant considers the issue so obvious that it doesn't even deserve arguement, writing "“The nominal definition of truth, that it is the agreement of [a cognition] with its object, is assumed as granted.” (Critique of Pure Reason, 82). The Correspondence Theory has endured over the centuries, in part because it is seemingly so axiomatic, so elegant in its simplicity.

According to the CTOT, a statement is true iff (if and only if) it corresponds to reality. If I make the statement "Dr. Donougho is the Professor of my Philosophy class," this statement is true iff Dr. Donougho is, in fact, the Professor. If anyone else is the Professor (or if I don't have a Professor), then my statement is false. According to most interpretations of the CTOT, my statement would be false even if I have every reason to believe that Dr. Donougho is the professor, but he is not in fact because some impostor has taken his place. One's context of knowledge is irrelevant. Please keep this in mind, as we will have reason to return to it in a moment.

The Correspondence Theory of Truth is an alluring philosophical principle on many levels. At root, it is an epistemological expression of axiomatic metaphysical principles. Two such axiomatic principles in Objectivism are the Law of Identity and the Primacy of Existence. The Law of Identity states that "A is A," or "a thing is what it is." The Primacy of Existence states that "existence has primacy over consciousness," implying that consciousness has no effect on the identity of entities. If one accepts these metaphysical principles, then the task of man becomes to discover the identity of the world around him. His consciousness does not have the capacity to create, only to identify reality.

The Standard of Omniscience

The biggest problem with the common interpretation of the CTOT is that it sets a standard of omniscience for truth, making certainty impossible. If we accept that certainty is impossible, then we leave the discipline of philosophy open to skeptics (who will gleefully agree that reality is unknowable) and mystics (who offer a supernatural source for certainty).

As mentioned earlier, one's context of knowledge is deemed irrelevant when determining the truth value of his statements. If I say that "Dr. Donougho is my Professor," then my statement is false even if I have every reason to believe that it is true. These kinds of situations are not uncommon. The history of man is marked by an ever-expanding degree of knowledge about the nature of reality. In the realm of science, when new data is discovered that contradicts old theories, then the old theories are discarded, and new ones devised. Does this mean that the old theories were always false? How can we ever be certain that our theories will not be contradicted by new evidence at some point in the future? This creates a problem. If truth is determined without regard to context, then one's context must be all-encompassing (i.e., omniscient) to make a claim of certainty. An omniscient standard of truth is incompatible with Objectivism.

Knowledge As Contextual

Objectivist epistemology lays the foundation for a bridge between subject and object, and the reconciliation between the CTOT and certainty. Objectivism states that absolute certainty is possible within a specified context of knowledge. Any statement made by a human being necessarily implies the preamble "within my context of knowledge." This preamble is necessarily implied because man, by his nature, is a being of limited consciousness. He is not omniscient.

For example, Newton's Laws of Motion are true, and will always be true, given Newton's context of knowledge at the time. Einstein has access to better technology and higher levels of mathematics, and was able to expand man's understanding of Physics. He discovered new data that could not be explained by Newton's Laws, and he was able to construct a new theory which did account for the data. It would be false for Einstein to state that Newton's Laws will always always be true regardless of context, but Einstein could agree that, give Newton's context of knowledge at the time, his theories are still true.

The contextual nature of knowledge allows man to continually expand his understanding of the world around him, while at the same time avoid being paralyzed by uncertainty. The rational man forms principles on the basis of evidence, and treats them as absolute unless and until he discovers new evidence that would require him to restructure those principles. This methodology can be applied to ethics, politics, and esthetics, as well as science.

The Implied Conditional of Absolute Principles

Another helpful way of viewing absolute principles is in the form of a logical conditional. One can assert "If there are no aspects of reality outside the context of my knowledge which come into play, then principle X will always hold true." If one has properly formed his principle, and integrated it with all of the evidence available to him, then this statement will always be true, forever and ever.

One can view this logical conditional the same as any other, in terms of truth value. We can represent the statement as:

p-->q

Where

p = there are no aspects of reality outside the context of my knowledge which come into play

And

q = principle X will always hold true

Note that the truth value of this statement follows the same format of any other conditional statement. If 'p' is true, then 'q' must be true. If 'q' is false, then 'p' must also be false. And if 'p' is false, then the truth value of 'q' is indeterminable.

Let's return to Newton to demonstrate this method in action. Newton makes the claim that "if there are no aspects of reality outside the context of my knowledge which come into play, then The Laws of Motion will always hold true." If Newton acquires evidence that his Law of Motion does not apply to a particular case (for example, when entities approach the speed of light), then he knows that an element outside his context of knowledge has come into play. Now, he is challenged to integrate the new data with his old principle, modify it, or discard it favor of a new one. Note that, while 'p' and 'q' in the above example can have varying truth values, the logical statement p-->q is itself an absolute principle.

A New Perspective on CTOT

I have argued that theories and principles can be held with absolute certainty within a specified context. How then can one marry this with the CTOT?

When one formulates a theory based on evidence he has gathered, and integrates it without contradiction into the whole of his knowledge, then his theory does indeed correspond with reality. He has properly identified a relationship between his consciousness and some specified aspect of reality. It is important to keep in mind that a conceptual consciousness is an entity in reality, and an understanding of that consciousness is an instance of correspondence. One's integration of data into concepts and principles corresponds to the reality of his conceptual consciousness, and the contents therein.

One will never be able to step outside of his consciousness and make propositions based on information unavailable to him, but that does not mean that his limited theories do not correspond to the aspects of reality within his contextual range. It is unfair, and indeed irrational, to demand that man define "truth" in terms of the metaphysically impossible, i.e., omniscience.

The Importance of Semantics

An understanding of epistemological concepts is critical to maintain the integrity of philosophy as a discipline. In academic circles, classrooms, journals, online discussion forums, and even private philosophical conversations, the defenders of rationality must insist on the precise use of epistemological terms like "truth" and "certainty." The reason why there is so much misunderstanding about the proper application of the CTOT is that many just assume that the CTOT requires a standard of omniscience. This misunderstanding must be identified, clarified, called out, and rooted out of existence.

When engaged in discussion with an intellectual opponent, especially on issues of epistemology, take care that your opponent is not demanding that you step outside your context of knowledge in order to make a claim of certainty. A flaccid refrain of rationalistic "what ifs" does not take the place principled, logical, evidenced-based discussion. In the of spirit of Socrates, define your terms, and challenge your opponent to define his.

Communication

I must anticipate a challenge to my argument with respect to communication. If all knowledge is contextual, and each individual is working from his own context, then how can one defend against the charge of relativism? Is it reasonable to assume that one can effectively communicate with other individuals, and that the ideas being communicated are understood in the proper context?

My answer to this last question is 'yes!', given that one goes to the effort of defining his terms. This is why the precise definition of philosophical terms within one's own mind is so important. Individuals may define concepts in subtly different ways, so when a misunderstanding arises, the first step is to make sure that participants in a discussion understand each others' context of knowledge with respect to the issues being discussed. Misunderstandings are always possible, but can be limited through proper discourse of ideas.

Also, while individuals may have different contexts of knowledge at any given time, as men we all have access to perceptual concretes. Another method of clarifying interpersonal communication is to logically reduce ideas as close as possible to the perceptual level.

Looking Forward: A Complete Theory of Induction

There are many challenges in the field of epistemology that must still be overcome. The most important of these, in my view, is a more complete theory of induction. Logical induction is man's method of forming rational principles based on evidence of the world around him. Ayn Rand presented a powerful theory of induction with respect to concepts in her Introduction to Objectivist Epistemology, but some questions remained unanswered.

First, what degree of evidence is required to form a (contextually) absolute principle? In Objectivism - The The Philosophy of Ayn Rand, Leonard Peikoff discusses degrees of certainty with respect to principles (p67). He classifies theories on a graduate scale, from "possible," to "probable," to "certain." But what delineates these degrees of probability? More importantly, at what point can one settle on any given principle as "certain?" One clue offered by Peikoff is that in order to settle on a principle, "all evidence points in one direction, there is no evidence in any other direction, and no contradictory evidence." This idea is helpful, but insufficient.

If philosophers can agree on the CTOT and proper standards of rational discourse, then I believe these questions are answerable within our lifetimes. The only way to defeat the specter of skepticism in the academic community is through our consistent and determined effort to define our terms in philosophy with the most rigorous logical integrity.

<!--[if !supportEmptyParas]--> --Dan Edge<!--[endif]-->

View the full article

[DavidOdden]

I have a number of comments that I'll offer on this essay tomorrow, but for the moment I'd just like to say that I appreciate this serious taking on of a very bothersome philosophical issue, one that used to bug me about Objectivism for a few years, especially from my professional perspective. The central problem is the idea that "a statement is true iff (if and only if) it corresponds to reality". The problem as I see it hinges on the notions of "statement" and "correspond", two ill-defined notions used by philosophers to avoid saying what really is the case. If the concept truth is explained in terns of "grasping" i.e. comprehending or in new-speak "transducing", and specifically a mind grasping fact, then many classical problems are avoided. Sentences are produced by minds, with reference to social rules. Most problems of "context" as applicable to truth are about the social aspect and not about the actual "correct grasp" aspect.

But, more in the morning. Nice essay, in general.

[aleph_0]

Interesting essay, I like it. I would say that you should be careful to avoid Kuhnian "paradigm shift" ideas when you talk about how "Newton's Laws of Motion are true, and will always be true, given Newton's context of knowledge at the time." I suggest adding something that makes the case that we can now say, "I know with certainty..." because the assumed qualifier is, "Within *my* context of information, at *this* time"--i.e. that the assumed qualifier is extremely indexical. When talking about Newton, you cannot say that his laws are true and always will be true, period, because the assumed qualifier is "within my context of information, at this time". If you were to say, "Within my context of information at this time, Newton's laws are and always will be true," that is a false statement. So when talking about somebody else who has a distinct knowledge context, you must always specify whether you are talking about your (the speaker's) context, or his (the subject's) context. In this way, paradigms do not shift, but rather the individual claimant talks about what his state of awareness necessarily requires of him--either to believe in Newton's physics or Einstein's.

One problem, pervasive in the philosophy of science, that you will face--and I have no idea how to solve it--is the issue of tack-on claims. For instance, you could have the claim, "Based on all of my empiricle evidence, x must be true," and you symbolize it P --> Q. Yet take the sentence "god is real" to be symbolized by G. Then the statement (P v G) --> Q is true as well, and so is the English version, "If either of the following hold: My empiricle evidence is sufficient to grasp the unqualified truth or god is real, then x must be true."

Also be careful to note in which context of knowledge you intend when you say, "q = principle X will always hold true," or whether this is supposed to hold independently of any context. The same for when you go on to talk about "Note that the truth value of this statement follows the same format of any other conditional statement. If 'p' is true, then 'q' must be true. If 'q' is false, then 'p' must also be false. And if 'p' is false, then the truth value of 'q' is indeterminable."

One question I have is, when you say, "When one formulates a theory based on evidence he has gathered, and integrates it without contradiction into the whole of his knowledge, then his theory does indeed correspond with reality," did Newton's laws correspond with reality? Talk of "reality" seems to imply claims without respect to any context.

[DavidOdden]

Some have expressed dissatisfaction with an unqualified acceptance of the Correspondence Theory

I'm a dissatisfee, of a particular kind, namely one who finds many versions of CT to be vague and even wrong about what things "correspond". The typical contemporary presentation goes like this: 'The statement "X" is true iff, in fact, X', where X is a declarative sentence of the language under investigation. From what I can tell, "X" is simply the linguistic form and X is something else, but I've never understood what it is supposed to be. Since CT is typically taken to describe a relationship between an aspect of consciousness (for example a sentence, proposition, or statement) and reality (fact), then we have to conclude that X is a fact (and not a description of a fact), which is what "X" is supposed to be.

Here are some examples of consciousness / existence relations:

• "That is a dog" iff that is a dog.
  
• "Dat lea beana" iff dat lea beana.
  
• "Det er hund" iff det er hund.
  
• "C'est un chien" iff c'est un chien.
  
• "Haann aba 'i" iff haann aba 'i.
  
• "Hiyo ni mbwa" iff hiyo ni mbwa.
  
• "Sqwebay ti'ilh" iff sqwebay ti'ilh.
  
• "Это собака" iff это собака.
  

If it's not obvious, these are synonymous descriptions of the same fact of reality across a variety of languages. The contemporary CT approach in effect denies the relevance of existence and favors consciousness as the philosophical primary by trivializing the existence side of the correspondence. Modern CT favors consciousness-reification saying that on the reality side, we simply have the mental product taken out of quotation marks.

The Objectivist perspective on truth is different: it states that "truth is the product of the recognition (i.e., identification) of the facts of reality". Applied to dog sentences (or anything else), you start with reality. There are existents out there, whose nature can be grasped by any man and integrated into a concept. The physical nature of the cognitive object, the dog-concept, will be different in my brain from the physical representation in my wife's brain or my neighbor's brain, but the referents of the concept -- the meaning, the concretes identified by the concept -- are the same. Then to say the true sentence "That is a dog" or the various translations into other languages, we must recognise one aspect of reality -- the existents linguistically labeled as "dog", "beana", "hund", "aba" etc. -- and the relation of the concept to another aspect of reality -- a particular concrete on the floor here, in particular that the concrete existent is subsumed under the broader fact known as "beana" (inter alios).

In other words, Objectivism avoids the absurd multiple-realities position that arises from modern CT which sees existence as a projection of the mind (i.e. a sentence taken out of quotation marks). By taking existence to be primary and consciousness (including recognitions such as "truth") to be secondary, we maintain the single- and objective-reality theses (which I take to be fundamental philosophical desiderata).

Now we come to the point where I potentially disagree with what you say (depending on what you intended), namely the nature of the "context" problem. As you say, under CT, "One's context of knowledge is irrelevant" and thus "The biggest problem with the common interpretation of the CTOT is that it sets a standard of omniscience for truth, making certainty impossible." Clearly, the Objectivist account of truth cannot eschew knowledge: our theory of truth is that truth is the product of a product of a mind grasping fact, which is knowledge. At the same time, it is wrong to open the floodgates and make the Objectivist theory of truth become a variant of justified belief theory. In reality, not all crows are black, but many people are completely unaware of that detail about reality (AFAIK, all crow species in America are in fact black). Thus the statement "All crows are black" is not truth, it is false, and a theory that equates ignorance with truth is a mistaken (indeed, false) theory of truth. Even stranger, some humans believe that penguins are mammals (I was surprised to encounter such a belief, myself), and clearly we don't want to say that "Penguins are mammals" is true. In other words, we don't want "contextual knowledge" to be an epistemological get out of jail free card. What we will probably want to conclude is that Newton's law of gravity was and is false, though there might be a related revised law (which I've never seen articulated) which is true.

Now we come to the concrete application of the idea of contextuality to scientific theories:

In the realm of science, when new data is discovered that contradicts old theories, then the old theories are discarded, and new ones devised. Does this mean that the old theories were always false?

Yes, it does mean that, if in fact the new data contradicts old theories. The bare proposition Fg=(m1*m2)/r**2 is and always has been false, although we were not squarely aware of that falsity until the 20th century. The proposition was not certain (in the Peikovian sense) which is why it is false. There might be an alternative proposition that is true, if conditions are put on the values m1, m2, r, but I'll leave that to a physicist to do. The point is that the contradiction arises because of the particular expression.

Returning to the crow problem, when a person learns about corvus albus and thus encounters an apparent contradiction, what has happened is not that the person's knowledge context is expanded from "All crows are black" to "All crows are black, or not"; rather, the person's knowledge of the term "crow" has expanded. The person then learns that "crow" doesn't just refer to corvus caurinus (that was my childhood version of crow), but refers to something broader, and the thing that the person grasped correctly but mislabeled is the relationship between the existents (a particular species of crow) and their properties (color). The person's error was essentially social, in misconstruing what the word "crow" means, in the wider context of English-speaking society.

You ask:

How can we ever be certain that our theories will not be contradicted by new evidence at some point in the future?...If truth is determined without regard to context, then one's context must be all-encompassing (i.e., omniscient) to make a claim of certainty.

I suggest that the proper focus for concern is in correctly identifying the proposed statement. (I endorse your recommendation for being explicit in definitions, probably even more strongly than you do). That is, it is necessary to recognise the objective aspect of meaning, and also recognise that usually, people are wrong in their overt statements about objective meaning. We must always avoid the fallacy of redefinition, where people can declare "if we redefine "dog" as 'those animals known scientifically as corvus albus' then my argument is true". We must also avoid the intrinsic-meaning fallacy, where dogs are called "dogs" because of some intrinsic property of dogs, or the sounds of the word "dog". A large portion of the "context" problem is exactly about establishing the socially correct meanings, to avoid concluding that "crow" actually means just corvus caurinus. If you are aware that "crow" refers to a large family of birds and that you have validated your crow-color theory for only one member of the family, then you cannot be certain of your conclusion because there is (fairly) well-known conceptual and empirical evidence that would point in the direction of an alternative hypothesis.

Now here we come to a very important fact that, in my mind, contextualizes context especially as applied to ordinary knowledge versus the desiderata of scientific knowledge:

Any statement made by a human being necessarily implies the preamble "within my context of knowledge." This preamble is necessarily implied because man, by his nature, is a being of limited consciousness. He is not omniscient.

I don't think this identifies the problem correctly (though I haven't thought about exactly what 'omniscience' implies, as opposed to 'infallibility'). I would, rather, conclude that the problem stems from fallibility and efficiency. A statement from a human implies some knowledge context for a reason: it is inefficient to make explicit that knowledge. There are very many things that you implicitly know and depend on in communicating with other people, for example you don't start a conversation about dogs by laying out the definition of "dog", or the definition of "this", and unless you're Bill Clinton, nobody explicitly defines the verb "be". To do so would make communication virtually impossible, and would thwart the cognitive purpose of having a conceptual faculty. I conclude then that most applications of the notion "contextual knowledge" depend more on a failure, namely a failure to be explicit when you should be, or a failure to operate according to social norms for communication, the latter being especially sensitive to real-world context (that is, who you are talking to).

Here's an example of contextual variability in meaning. In virtually all (but probably not actually all) philosophical, mathematical and legal circles, "all" means "all, excluding none". In ordinary useage, "all" is not interpreted so stringently, and universal statements will usually be considered true if there is one counterexample. (I'm not condoning that use, just pointing out that it exists). If you fail to conform to the contextually appropriate meaning of a word, you're not necessarily suffering from a lack of knowledge, you're suffering from a failure to use the knowledge that you do have. That is, you failed to define your terms explicitly so that the other party understands the difference between what you are claiming and how they would understand your statement, under normal rules of interpretation. (It's of course also possible to simply be ignorant, for example, the person who thinks that penguins are mammals is not confused about what "penguin" means or what "mammal" means, s/he is just wrong about the nature of penguins -- it is still not contextually true that penguins are mammals).

The point that I want to emphasize most strenuously is that Objectivism does not adhere to a justification theory of truth, which is what can result from over-application of the notion "contextually true". Newton's failure to grasp the falsity of his law is not a sign of moral weakness on his part, it arises from many sources such as his religious beliefs, the nature of scientific instrumentation at the time, and so on. We have to come to terms with the fact that a person can have an honest belief in a false statement. Newton was and is wrong: that doesn't mean that he should have known that (since, after all, ITOE and OPAR were written hundreds of years later).

[dan_edge]

Aleph,

Thanks for the comments. I agree that this essay could use a lot of expansion and clarification. It is as long as it was supposed to be for my class, and no longer. Your point is well taken about making sure always to specify context clearly.

You asked about this statement: ""When one formulates a theory based on evidence he has gathered, and integrates it without contradiction into the whole of his knowledge, then his theory does indeed correspond with reality."

I thought that the rest of that paragraph makes clear the sense in which ideas correspond to reality: "He has properly identified a relationship between his consciousness and some specified aspect of reality. It is important to keep in mind that a conceptual consciousness is an entity in reality, and an understanding of that consciousness is an instance of correspondence. One's integration of data into concepts and principles corresponds to the reality of his conceptual consciousness, and the contents therein."

I'm not sure what is unclear about this paragraph. Do you have a problem with the terminology? Do you dislike the phrase "correspond with reality" in general?

Thanks for the feedback!

--Dan Edge

[dan_edge]

David,

Whoa. That's heavy. I will have to digest your post for a while and come back to it again. My initial thoughts:

Let's consider the Newton example further, because I don't fully understand your position. Let's make the example more clear and assume that Newton did not commit any dishonest errors, rigorously followed the scientific method, and in general acted like a good Objectivist. In this case, I argue that Newton's Law of Gravity is in fact true and will always be true. It is a fact that, given a certain context, gravity causes objects on earth to accelerate at 9.8 m/s(squared). Given that one is using a measuring instrument which provides accuracy to a tenth of a meter, this theory absolutely holds true today. Newton's Laws are still taught in science class as true theories within a specified context, and I think that's appropriate.

When Newton claims that his Law of Gravity is true given his context of knowledge, he is referring to the same contextual restrictions on the Law that we place today: on planet Earth, using measuring instruments with a certain degree of precision, factoring out wind, etc. His concepts are referring to the same existents and abstracting away the same factors. The one thing that we have over on Newton is that we can add other qualifications to the Law that he was unaware of: that the falling objects are not moving at very high speeds, approaching the speed of light, and that we are not using measuring instruments with .00000000001 (or whatever) degree of precision. I don't think this makes his theory false, because, assuming he was a good Objectivist, he does not state his theories like this: regardless of context, objects will always fall at 9.8 m/s(squared).

As you anticipate, there is a restatement of the Law of Gravity which accounts for relativity. (I don't recall it offhand, but it's essentially the same formula multiplied by (a bunch of stuff divided by the speed of light)). This formula is only used in certain cases, like when objects are moving very fast with respect to one another or when one is using very precise instruments. The precision of this formula is based on the precision of our measurement of the speed of light, since c (representing the speed of light) is included in the formula. Certainly this would not imply that every statement of the Law of Gravity is false (or that every claim to certainty about a measurement of a falling object based on the Law of Gravity is false) when scientists learn how to measure the c with more precision? I know that's not what you were arguing, I'm just trying to be clear.

Considering these facts, what would you say is the truth value of Newton's Law of Gravity (for instance)?

It will take me some time to integrate what you said about concepts with my understanding of CTOT: "The physical nature of the cognitive object, the dog-concept, will be different in my brain from the physical representation in my wife's brain or my neighbor's brain, but the referents of the concept -- the meaning, the concretes identified by the concept -- are the same."

I think you're absolutely right to focus on this point, but I don't see clearly how to distinguish (or whether to distinguish) concepts from statements of fact. Perhaps a grammar course is in order for me? My fiancee's dad has Peikoff's Grammar course, and has recommended it highly to me. Have you (or has anyone else) listened to this course? Do you think it would help me better grasp the relationships between concepts, declarative statements, and truth values?

Thanks for taking the time to respond at such length.

--Dan Edge

Edited April 28, 2007 by dan_edge

[DavidOdden]

Let's make the example more clear and assume that Newton did not commit any dishonest errors, rigorously followed the scientific method, and in general acted like a good Objectivist.

Man, you are asking a lot. We know that in fact Newton was very religious (not a good Objectivist), and that he was a bit of a nominalist. Let's use OPAR ch. 5 as our characterization of the scientific method (since it is, even if sometimes scientists don't act that way). Now also let us be clear on what the Law of Gravity says: Fg=(m1*m2)/r**2. Not, Fg=(m1*m2)/r**2 for a specified range of m1, m2, r, but unconditionally Fg=(m1*m2)/r**2. (I recall that there is evidence, which Stephen Speicher presented in a series of posts on HPO a few years ago, to suggest that although Newton presented this law as a universal law, had second thoughts about the universality of The Law. This would be important to understanding Newton as a man / scientist, but Newton's Law is conventionally taken to in fact be an absolute law, not a law with guard conditions on it -- I've suggested that such a law might be possible and true, though not conceptually superior to modern understandings of gravity).

In other words, if you say "Newton's Law is true subject to the following conditions, that when m1 or m2 is very small then.... or m1 are very large... then we predict these values up to N decimal places", that might describe a fact, but that isn't Newton's Law. A true law would include the context.

When Newton claims that his Law of Gravity is true given his context of knowledge, he is referring to the same contextual restrictions on the Law that we place today: on planet Earth, using measuring instruments with a certain degree of precision, factoring out wind, etc.

Wait, huh? I don't recall Newton saying anything at all about restricting the law to his context of knowledge -- he meant is as applying to all contexts (hence "universal law of gravitation"). He himself was aware of a conceptual basis for doubting his law (the action at a distance problem), and he recognised that, in modern parlance "it's just a model" without any understanding or implications of causality (we know that in fact all true laws embody causation). I should also point out that theory is best tested off of Earth, since on Earth, there are zillions of alternative hypotheses that are empirically as plausible (since the mass of the Earth is the proverbial elephant in the room). But when you go to the motion of the planets, the empirical support simply does not rise to the level of certainty, thus actual knowledge. The law was, in the empirical context of the day, probable to a considerable degree, but it did not rise to the level of proof.

BTW, can you locate a copy of his actual law (i.e. text copied from the Principia)?

Considering these facts, what would you say is the truth value of Newton's Law of Gravity (for instance)?

What exactly are you referring to when you say "Newton's Law of Gravity"? Fg=(m1*m2)/r**2 is false. Something else may be true -- I think some Canadian guy has a workable version. Let's first define the expression "Newton's Law of Gravity".

I think you're absolutely right to focus on this point, but I don't see clearly how to distinguish (or whether to distinguish) concepts from statements of fact.

Well, a trivial example is that "dog" is a concept, and "My dog is snoozing on the couch at 4:18pm, April 28, 2007 is a (singular) statement of fact. A concept has stuff on common with the broader notion of "understanding", but concepts specifically reduce multiple mental units to a single symbol (this is the whole cognitive economy thing). A concept must be represented by a word (or analytically equivalent symbolic unit -- in the biz, we call them "morphemes" since concepts can be represented by roots that aren't full words). But not every word represents a single concept -- this is the phenomenon of polysemy. That is, some words are imprecise and require contextual disambiguation to make clear which sense of the word you intend.

[dan_edge]

David,

What exactly are you referring to when you say "Newton's Law of Gravity"? Fg=(m1*m2)/r**2 is false. Something else may be true -- I think some Canadian guy has a workable version. Let's first define the expression "Newton's Law of Gravity".

I was trying to use a historically-based example and assume the best case scenario on Newton's part, but maybe that was unclear. What I mean is: for the sake of argument, assume that Newton was an Objectivist and perfectly formulated his Law of Gravity based on the data available to him at the time. He presents a formula and explains conditions under which it holds true. So assume Newton's Law of Gravity is Fg=(m1*m2)/r**2 along with an explanation of what the terms mean and which factors to abstract away (like wind or other forces). I argue that this theory is true, and it is still true. Fg=(m1*m2)/r**2 always, always works within a certain range of precision and assuming m1 and m2 are not moving at very high speeds.

I just don't see how one could say that, given this example, Newton's Law was false when he uttered it. And it doesn't make sense to say that was true when he said it, and false now. It makes more sense to say: it is true given his context of knowledge at the time, but false given a different (broader) context.

--Dan Edge

[DavidOdden]

It makes more sense to say: it is true given his context of knowledge at the time, but false given a different (broader) context.

How about the theory that the Earth is flat: was that true given the context of knowledge at the time that the claim was believed and false given a broader context? I'm trying to see how your version of truth is different from justified belief theory.

I would hold that an Objectivist scientist should state the conditions on their scientific claims explicitly, so that if you construct a law and test it against 100 existents, you law isn't "contextually true" because it was true for the cases you considered. I assume (perhaps wrongly) that you agree on that: then what distinguishes proper contextual expansion of knowledge, vs. simple error.

[dan_edge]

David,

What exactly is "justified belief theory?" The way you write about it, it seems to be a specific theory, but I haven't heard of it. I tried to look it up through google and wikipedia, and came up with mixed results. There appears to be no consensus view of what a "justified true belief" is. Do you have any references to what you mean by "justified belief theory?"

Thanks,

--Dan Edge

[DavidOdden]

What exactly is "justified belief theory?"

Ah, well, I am not totally sure. It may be that it's disregarded in favor of JTB theory, or even, it may be a figment of my imagination which came up in discussions on HPO over the Objectivist theory of truth. At one point I thought that "Justified Belief" theory was a version of ancient Greek epistemology which Plato didn't like because it allowed the possibility that you could know something that wasn't true. I got this impression from slogging through Theaetetus, and obviously I could have totally misunderstood or misremembered this.

The problem is that JTB can't be a theory of truth, it has to be a theory of knowledge, because to say that truth is a justified true belief is, well, obviously circular. But then knowledge is subordinate to truth, and we can (under JTB theory) know a fact but not know that we know the fact. JTB theory undermines the value of knowledge, since only an omniscient being can distinguish knowledge from error, if "true" is part of the definition of knowledge. And of course, JTB theory has no account of what truth actually is, separate from CT. What I wanted to emphasize is that it is crucial that "truth" not be at all evidence / justification-dependent.

[dan_edge]

David,

What I wanted to emphasize is that it is crucial that "truth" not be at all evidence / justification-dependent.

I agree; this is the crucial point. The route I took in my paper to integrate CTOT with Objectivism was to stress that "truth" is an epistemological, not a metaphysical, concept; and that truth is contextual (evidence/justification-dependent). A claim to knowledge is true within a specified context, and that context is determined by evidence gathered from reality. This is my understanding of the Objectivist approach. What is the flaw here, in your view?

Let's consider the "flat earth" example. If a scientist living in the 21st century claims that "the Earth is flat," then it's easy to say that his statement is false. But if a 5th century scientist made the claim, "within the context of my knowledge, the Earth is flat," then what is the truth value of his statement? One could make the case that this is a true statement. The scientist's claim is epistemological, not metaphysical. He's making a claim regarding a logical relationship among existents, based on evidnece. He is claiming, in essence, "based on all the data I have gathered, these units are related to these other units in a certain way, and all the evidence at my disposal confirms this conclusion." This statement is fact and corresponds to the relationship between the scientist's mind and reality.

--Dan Edge

[DavidOdden]

The route I took in my paper to integrate CTOT with Objectivism was to stress that "truth" is an epistemological, not a metaphysical, concept; and that truth is contextual (evidence/justification-dependent). A claim to knowledge is true within a specified context, and that context is determined by evidence gathered from reality.

I think I see the problem.To quote ITOE p. 48, "Truth is the product of the recognition (i.e., identification) of the facts of reality". A proposition is true if it correctly describes the facts, and a person who grasps a true proposition has knowledge. Truth is not purely epistemological, rather it is both epistemological and metaphysical -- it is a particular relationship between a consciousness (the epistemological part) and the fact that it is aware of (the metaphysical part).

You are also raising another, more complex question, namely a quasi-normative one about a claim to knowledge. It's not that I object to talking about claims to knowledge, but I think that any mention of claims to knowledge have to be put on hold until the question of "true" and "truth" are sorted out. My opinion is that mixing arguments about claims about knowledge with arguments about what knowledge is will lead us down the dark JTB path that made Gettier famous.

But if a 5th century scientist made the claim, "within the context of my knowledge, the Earth is flat," then what is the truth value of his statement? One could make the case that this is a true statement.

I think if I were presented with such a claim by a scientist, I'd ask him what exactly he thought was claiming. To paraphrase the claim, he seems to be saying "The facts which I have considered tell me that the Earth is flat". As such, the claim is primarily about the scientist's justification for believing the Earth to be flat -- that's why people hedge, saying "As far as I know...", "All of the evidence I know of says...".

Let's compare two statements -- "As far as I know, all crows are black" and "All crows are black". Also, let's presume that the person saying these things is someone who isn't me, i.e. a person who is unaware of non-black crows. The first statement is true and also subject-dependent -- it's the kind of statement where it really is apt to say that "it's false for me, but true for you". The second statement is a statement about objective reality, and not a statement about personal knowledge.

People commonly do intend to say "As far as I know, X" but all that comes out is "X". This omission makes a huge difference, because an actually false proposition can be embedded within the scope of a knowledge claim, where it certainly can be the case that a person doesn't know that he's wrong. The principle that scientists have embraced, for their claims, is that the claim should be explicit. If a scientist claims that a certain relation holds for masses between N and M, that means that the relation must hold for all masses between those values, and if you can show that the relation does not hold for some mass between N and M, the claim is simply false. Of course no scientist wants their claims to be disproven, but no scientist worth their salt would defend their claim by saying "I was right, as far as I knew, and my claim does not hold for anything that I did not know about at the time". That is, scientific principles are principles, and not lists of observations.

We distinguish between "being wrong" and "being in the wrong". A person who proposes a scientific principle in a particular form may be shown to be wrong, that is, his principle is not true because it does not correctly describe reality. But his method may be rather sound; so we don't castigate him for being wrong. But a scientist who employs unsound methods does merit castigation, because he should have known better. One (unforgiveable) sin of scientific method is claim-weaseling -- if you say "all", then you mean "all" and not "all that I knew of". Since the point of doing science is to make true claims, over-generous dispensing of contextual forgiveness encourages claim-weaseling, in that it allows an actually false claim to be considered to be true. What we want is the actually true claims, the claims that correctly describe reality, and not claims about what people think they are justified in believing.

[dan_edge]

David,

You make some good points. I'll be going over Objectivist epistemology again in the OPAR class next semester (fall?) in the OAC. I'll likely revisit this issue when it's fresh in my mind.

Thanks,

--Dan Edge

[aleph_0]

I find it very interesting that I just read an article in the philosophy periodical Think, titled "An Argument for Relativism" by Max Kolbel which argues nearly this very same thing, but only argues it for the relativity of tastes. Thought you might like to know.

...

Oh, and one further question that I have--and forgive me if this has already come up--is whether the context itself must be understood in a particular context. Tthat is to say, when you claim, "I have such-and-such a context," do you need yet another context in order to make that claim (since all claims must be made with reference to a particular context)? Also, what about the claim about the relationship between the context and the truth that it implies? For instance, it seems necessary for you to make the argument that, in all contexts, one ought to recognize how Newton's context required him to believe in Newtonian physics.

Edited May 3, 2007 by aleph_0