Colin and I have been having a disagreement by email which prompted me to put some thoughts down here in a more orderly fashion than I did in my emails.
Let X be any statement that could be considered factual. (Leave ‘factual’ as primitive for now.) Consider four phrases that could be appended to X:
a. It is true that
b. It is the case that
c. It is believed that
d. It is known that
(a)X will be an abbreviation for “It is true that X”, (b)X is an abbreviation for “It is the case that X”, etc.
Excluding performatives, in any context where some statement (a)X, (b)x, (c)X, (d)X is appropriate (requested, desired, etc), the statement of X without a preceding phrase will also suffice. “Suffice” here means “not result in communicative misfire.” I think this also hold in at least some cases with performatives but I’ll leave that for now.
Here are two examples with different statements for X.
[I] Take X for this example to be “The soup has nuts in it.” Let’s say someone with nut allergies wants to know if they are safe eating the soup I’m serving for dinner. They ask me “Is it true that X?” I can answer with any of (a)X, (b)X, (c)X, (d)X, or X without a preceding phrase. In this context, X and the four compounds of phrase plus X are equivalent.
[II] Take X for this example to be “The product is harmless.” Let’s say I sold a product which was found to be harmful. I am on trial to see if I did so deliberately. The lawyer asks me “Do you believe that X?” I can answer with any of (a)X, (b)X, (c)X, (d)X, or X without a preceding phrase. In this context, X and the four compounds of phrase plus X are equivalent. (Saying “The product is harmless” would also here indicate that I do not assent to the finding that the product is harmful.)
In example [I] the issue at stake is that of determining an external state of affairs so to speak - whether or not the soup has nuts in it. In example [II] the issue at stake is that of determining an internal state of affairs, so to speak - whether or not I acted with deliberateness in selling a product which has been found to be harmful. Whether the issue at stake has to do with an internal or external state of affairs is not contained in (a)X, (b)X, (c)X, (d)X, or X. Rather, the issue at stake is determined in the context in which the statement appears.
It would not make sense to say “(a)X but not (b)X” or the reverse. So, let (a)X and (b)X be considered equivalent. Let’s say “(a/b)X” is an abbreviation for “(a)X or (b)X.” Modify (c)X and (d)X to apply to one subject in the first person: “I believe that X” and “I know that X.” In that case, it would not make sense to say “(c)X but not (d)X” or the reverse. So, let (c)X and (d)X be considered equivalent. Let’s say “(c/d)X” is an abbreviation for “(c)X or (d)X.” I hold that (a/b)X does not differ in any context independent way from (c/d)X. That is, in at least some contexts (a/b)X is equivalent to (c/d)X, something that can be abbreviated (a/b/c/d)X. Examples [I] and [II] show this.
(One point of my and Colin’s disagreement is that I argue that one can not honestly assert “(c/d)X but not (a/b)X” or the reverse. Colin argues that one can. I’m not sure how to address this.)
There are some examples where these are not equivalent. I can think of two.
The first example deals with false beliefs, and has two types. One can find out that one has had false beliefs (this happens in time, a distinction between present and past: “I believed that X but X is not true” or “I believed that X but [now] I know that not X”) or one can encounter or know of some person or group with beliefs that are false (this is a distinction in terms of group membership “they believe that X but X is not the case” wherein the speaker does not belong the group indicated in “they). One can not, however, be a member of a group in the present which both holds (c/d)X but not (a/b)X.
The second example deals with uncertainty. Stating (c)X can be used to indicate that one is less sure than if one says (a)X, (b)X, or (d)X. Let’s say for the sake of argument that “I know” is equivalent to “I am completely certain.” This does not mean that there is a general difference in kind between (c)X and (d)X. Rather, the difference is one of degree: one can be very certain but not completely certain, for instance. One can also be completely certain but still be wrong. That one says “You believed X with complete certainty but you were wrong” but does not say “You knew X but you were wrong” strikes me as an idiomatic difference, not a difference between the substance of knowledge and belief.
These examples do not mean that there is an important context independent difference between (a)X, (b)X, (c)X, or (d)X. I believe that is means there is not a context independent difference between being-true, being-the-case, being-believed, and being-known.
The other big point of my disagreement with Colin, I think, is that I think he believes we can have truths which are not things which are taken-to-be true but are actually true. That is, that we can get outside the position from which we hold our beliefs and in some way verify them in a way which is not subjective. I don’t understand how that would work. I hold that “objective” can always be replaced with “subjectively considered objective” with little loss.
Simply put it is irrational to hold a belief consciously that you know or believe not to be true. That is, your beliefs have to align. When they don’t, you have a form of irrationality (psychology makes studies of when people accommodate new evidence to maintain false beliefs). You can have beliefs tacitly or unconsciously that contradict your conscious beliefs. Not sure if that helps
Comment by todd — December 26, 2006 @ 5:22 pm
Hey Nate–
This is a great distillation of our respective arguments… I’d like to respond to a couple of points, though.
1. In the example of the potentially-nutty soup, only a(x) and b(x) are equivalent. Saying d(x) is different than saying a(x) or b(x), though it offers equal certainty with respect to the nuttiness of the soup. You can’t say d(x) unless you know the truth of a(x) or b(x), so if you say you know that the soup is nutless, the person can trust you, and eat the soup. Saying c(x) is different. If you say you believe the soup is unnutty, they’ll probably ask you to check the ingredients to make sure.
The likely response of the nut-allergic would-be soup-eater to these statements indicate the differences between the at least 3 of these statements, and their respective non-equivalence. 1). The behavior of the nut-allergic would-be soup-eater in response to statements a(x) or b(x) does not take into consideration the one asserting a(x) or b(x). It presumes that it is a matter of fact that the soup is safe to eat. 2). The soup-eating of the nut-allergic would-be soup-eater to d(x) indicates that they trust the authority of the knower-of-the-nutlessness-of-the-soup. If they eat the soup, it is because they know or believe that d(x) depends on the truth of a(x) or b(x). This does not mean d(x) is equivalent to a(x) or b(x), but indicates that there is a relation dependence between them. 3). The nut-allergic would-be soup-eater’s response to c(x) does not or should not trust the authority of the believer-in-the-nutlessness-of-the-soup, because the believer could very well have no idea of the truth of a(x) or b(x). They might not d(x). And without the determination of that truth, or knowledge of it, caution is warranted on the part of the nut-allergic would-be soup-eater. Therefore, the statements are different and non-equivalent.
2. The example of the manufacturer of unsafe products serves to point out the difference between objective statements and subjective statements, as well as the non-equivalence of objective statements and statements held to be objective by some subject.
In this case, there are two different issues involved. One issues pertains to the simple fact that the manufacturer sold unsafe products. The statement “the manufacturer sold unsafe products” is valued according to a(x) or b(x), and once the truth of that statement is determined, we can state d(x).
Until a(x) or b(x) have been determined, one can regard the claim that “the manufacturer sold unsafe products” as a version of c(x), that is, something which is believed by a subject to be objectively the case. The problem is that this subject can hold c(x) regardless of whether or not the truth of a(x) or b(x) has been established. The manufacturer can be believed to have sold unsafe products without any good reason. You could believe that it was the manufacturer who sold unsafe products, when in reality the salesman modified the products, making them unsafe. In that case c(x) on the one hand and a(x) and b(x) on the other would be non-equivalent. One could render a fair judgment of the manufacturer only by deriving c(x) or d(x) from a(x) and b(x).
The manufacturer’s knowledge or belief with respect to the safety or unsafety of the products he sold is simply a different issue than a(x) or b(x). He might not have known that his products were unsafe, or he might have believed that they were safe. Neither changes the fact that he sold unsafe products. If he knew that the products he sold were unsafe, or believed that they were safe without testing them to make sure, then he’s not only a bad manufacturer, but guilty of reckless endangerment or negligence to boot. So, the issue of the knowledge and/or belief of the manufacturer in the safety or unsafety of his products is different from the fact that they are or are not unsafe. If he knew that his products were unsafe or believed that his products were safe without good reason to believe it, then he’s guilty of something over and above selling unsafe products. That difference points out the difference between the objective condition (having sold unsafe products) and the subjective condition (which considers the subjective state of the manufacturer when he sold the unsafe products.
Comment by Colin — December 29, 2006 @ 6:04 pm
hi Colin,
I’ll come back to this soon. I plan to try to distill some more of my thoughts from our email exchanges and put them up here. For now, one of the main sticking points between us seems to be the position of facts. I think the fact of the manufacturer’s having made a harmful product is not subject to verification (such that it can be demonstrated or known to be a fact) without something like beliefs entering into the picture. That is, objective conditions are reached via subjective conditions. This is not necessarily to say that the subject(s) create objective conditions, but rather that the assertion of objective conditions is itself an assertion by some subject. Gotta run.
take care,
Nate
Comment by Nate — December 29, 2006 @ 7:12 pm
“objective conditions are reached via subjective conditions. This is not necessarily to say that the subject(s) create objective conditions, but rather that the assertion of objective conditions is itself an assertion by some subject.”
I agree. I just think there has to be a way to establish that a,b, and d correspond to one another in a more-than-merely-subjective or more-than-merely-intersubjective fashion, such that knowledge can be more than a more-or-less-coherent system of beliefs. Recall that my main problem with belief is its independent variabiation with respect to a,b, and (maybe) d.
Comment by Colin — December 29, 2006 @ 8:52 pm
variabiation: variability, variation. Accident, not neologism.
Comment by Colin — December 29, 2006 @ 8:54 pm
hi Colin,
I understand that that’s the opposition you have. Part of where we differ I think is that for me you have a problem with an ineliminable condition, rather like having a problem with death (which is to say, I’m not necessarily happy that there’s not much more to knowledge than belief, I just think that’s true - a little ironic perhaps for me to assert here as this is probably more certainty than I should allow, strictly speaking). It’s also a little ironic that you “there has to be a way” is itself a belief rather than knowledge, I hope that’s not rude of me to point out. Gotta run again, soup’s on.
tchuss,
Nate
Comment by Nate — December 30, 2006 @ 12:02 am
I think it’s a little bit of a cheap shot to take my “there has to be a way” as a statement of a belief. I’m not offended, but I have to argue with it.
I suppose I would give a kind of transcendental argument to prove that my “there has to be a way” isn’t a statement of a belief. It would go something like this: If we know anything at all, then it’s possible to correlate knowledge with fact or truth. Even if we don’t know the means by which this correlation occurs, we know that it does occur, because we know that knowledge can’t vary independently of fact and truth. So, if we know, our knowledge has the same value as fact or truth, and is somehow determined by its correlation or correspondence with the values of fact or truth.
That argument isn’t a statement of belief, but a determination of the necessary conditions of knowledge. It’s valid, even if we don’t know the precise mechanism by which knowledge, truth, and fact are correlated. That mechanism requires a deduction, which is another kind of transcendental argument, and one I don’t claim to understand. That’s why I refrain from trying to define the mechanism by which knowledge, truth, and fact correspond.
Also, I don’t think belief is an ineliminable condition. I think it’s entirely eliminable. That’s why I keep insisting on the irreducibility of c(x) and d(x). Beliefs can change in ways that knowledge, truth, and fact can’t. We can (and, I think, should, though I don’t want that “should” to be mistaken for just another belief) become disillusioned with respect to beliefs that pertain to matters where fact, truth, or knowledge would suffice. That’s different than giving up on our hopes and dreams, but even there, the belief doesn’t do anything. It’s action and the realization of hopes and dreams that mean things, not the beliefs themselves.
Comment by Colin — December 30, 2006 @ 2:59 pm
hi Colin,
My apologies, it wasn’t intended as a diss or a cheap shot, I was in a rush and typed clumsily. That said, while it isn’t intended disrespectfully, I do think that the “there has to be a way” assertion has the status more like a belief than like knowledge. The conditional “if we have any knowledge” supports this. Presumably, we do have knowledge. I don’t think we could actually doubt everything or even most things. We can, for all intents and purposes, be said to know that we have knowledge. This isn’t a problem for me because I have lower standards for what constitutes knowledge than you do, I see it as much closer to belief. “We have knowledge” seems to be a founding statement from which our inquiry can begin such that there’s no reason not to treat it as true, but I don’t think it’s something you could conclusively demonstrate the truth of in the face of a radical skepticism. (In that case, one just ceases to talk with the skeptic.) That’s clumsy, let me try again.
It seems to me that your “there has to be a way” is a philosophic intuition which you’re working here to elaborate and defend. I can respect that, I’m doing the same with a different intuition. But neither is conclusively established and so is I think at this point more like a belief (or something wanted) than something known. Prior to establishing that a,b, and d correspond in the way you say must be possible, I don’t see how it can be said to be known that there has to be way to establish that correspondence. Prior to that establishment, it seems to me that it’s possibility can not be entirely determined. To say it could be would be I think essentially to say “there is such a way, it exists, just not in the present.” That would be to assert that at some future time conclusive evidence - some legitimate statement of type a(x) or b(x) - will be produced such that the statement “there is a way to establish a correspondence between types a,b, and d” can be shown to be of type d(x). That hanging of resolution upon the future is a conditional of the type that I called “belief2″ in my email.
take it easy,
Nate
Comment by Nate — December 30, 2006 @ 5:35 pm
For the sake of my own record keeping, here’s where my “belief2″ thing comes from, from a clumsy email I wrote to Colin -
“Clearly there’s a difference between three manufacturers of the same harmful product, wherein the first knows the
product is harmful, the second believes the product is harmful, and the third believes the product is not harmful.
There are two senses of belief available for the second and third. The first sense, belief1, is that which is taken for knowledge, based on false evidence or a failure of reasoning. Let’s say the second and third manufacturers believe the product is (not) harmful because of
something they read in their horoscope or in a dream about god. Or, say that the manufacturer is given a false document such that they are deceived (into believing the product is harmful in a different way than it really is, for the second manufacturer, and into believing the product is not harmful, for the third). They take these things seriously such that they believe they know that the product is (not) harmful.
The second sense, belief2, is something like a suspicion - something for which there is insufficient evidence such that one’s belief is probabilistic - “I might be wrong but I believe that…”. In this case neither manufacturer can be said to believe that they know the product is (not) harmful. Belief2 is something like a hypothesis or a conditional with a speculation on the existence of that which will make the conditional true. I think holding a belief2 goes something
like this, sticking with our example:
I believe the product is (not) harmful, but the evidence isn’t fully in yet. Once the evidence is fully in, my belief will be either confirmed or denied. I believe the evidence will confirm my belief.
This amounts to an if-then statement, with the assertion that the if indicates some uncertainty about or in the world and with the added assertion that once the uncertainty is resolved (or, if it could be resolved) then the if-then will be found to be true. I think the third sentence above, belief that the evidence will confirm the belief, is
important. For anyone to say “I believe X but I believe that tomorrow I will discover X is false” strikes me as absurd, akin to “I believe that X but I know that not X.”
Much of what I’m contending is that knowledge claims are always of type belief2 - with some exceptions, like logical rules etc, knowledge claims the denial of which would entail contradiction. Aside from these knowledge claims, any (we might say “any factual or empirical”) knowledge claim at time T1, such as “the product is not harmful”, is subject to being potentially verified or disproved at some future time
T(1+X). Should the claim be verified, we say “this knowledge claim is true” indicating a non-disjunction between v4 (”I know X is the case/true”) and v2/3 (” X is the case”/”X is true”). This assertion of verification is itself subject to potentially being later called into question, as are all statements. Should the knowledge claim be disproved, we retroactively change the status of the claim. We say
something like “he voiced an assertion of type v4 but was in error. What he should have said was an assertion of type v1, ‘I believe that X is the case/true’.” This is because in many contexts we do not allow for one to be said to know that X if X is not the case. That is, in many contexts statements of type v1 are allowable even if a corresponding statement of v2/3/4 is not currently allowable. Statements of type v4 are not allowable unless a corresponding
statement of type v2/3 is allowable. In that sense, “I believe” and “I know” are not identical in these contexts.
At the same time, I think the rules for allowability at any given moment in time involve at least implicit claims of type v1, which involve guesses about statements of type v2/3 (akin to statements of type belief2). After big shifts in beliefs, the rules for allowability change - heliocentric vs geocentric universe, atheism vs religion, etc. The new position still involves speculation, though (belief2) and
it is likely that at some point in the future some of the knowledge claims in the ensemble bound up with the rules for allowability will be show to be false such that we will retroactively say “there were not knowledge but belief”.
This still does not mean that there is an absolute disjunction between knowledge and belief. Any knowledge claim involves assertion of type v2/3. These types of assertions are often disproved. There is no way
of knowing, though, at the moment of making the claim that the claim involves assertions which subsequently be show to be false (such that one has only belief rather than knowledge).”
Comment by Nate — November 3, 2007 @ 3:48 pm