bd-from-kg

bgponder:

I had more or less abandoned this thread since it had moved far from the original topic. But your recent post provides a chance to clarify my position and answer some objections that have been raised against it.

1. What does “evidence” mean?

In spite of your impression to the contrary, “evidence” does not ordinarily mean “proof”. You don’t even seem to understand the definition you cited. Anything that makes a hypothesis more probable or plausible is a “ground for belief”, and it tends to prove it. The definitions found in many other dictionaries are clearer on this point. For example, the American Heritage Dictionary gives the following definition:

The Cambridge International Dictionary gives:

Obviously these definitions and examples show that anything that makes a proposition more probable or plausible is generally considered “evidence” for it.

In any case, when one of the terms used to frame a question is ambiguous, it is customary to choose the meaning that the person who posed it is most likely to have intended. Jamie_L was obviously not asking whether near-universal belief in a proposition proved that it was true, but was rather asking whether it was fallacious to claim that it made it more likely or plausible that it was true. (As he put it, those who make this argument claim only that near-universal belief is “supportive of the proposition in question” and rejects this argument by saying that “the fact that a particular belief is the norm really shouldn't matter”. ) So we should use a definition of “evidence” that captures his intended meaning.

2. On beliefs as evidence

It’s beyond strange that you should call my first post a “recovery”. What exactly was I recovering from?

Anyway, I thought I was reasonably clear. Near-unanimous opinion is “evidence”, but it’s not admissible evidence in a serious discussion. Why is this “ludicrous”?

Here’s another straightforward example. Say X is charged with murder. In the real world people charged with murder are guilty at least 95% of the time, so the fact that X has been charged is strong evidence that he is guilty. But it’s not admissible evidence. It would be outrageous for the prosecutor to argue that X is most likely guilty because he’s been charged, or for the jury to begin with a presumption of guilt on these grounds. But anyone not involved in the trial is perfectly justified in concluding that X is most likely guilty simply on the basis of the fact that he’s been charged.

3. On evidence and probability

It’s true that I believe that there are many propositions (such as “God exists” or “the half-life of Radon-222 is less than a week&#8221 for which it is meaningless to assign a probability. I made this point in the course of critiquing SingleDad’s attempt to characterize “evidentiary arguments” in general in terms of probability theory (e.g., Bayes’ theorem). I’m not sure that this is correct; the foundations of probability theory, and what it means in general to say that the probability of X is Y, is a vexed, controversial subject. As you point out, it’s true for any definite proposition that it’s either true or false, so in a sense the probability that it’s true is either 0 or 1. However, in some cases the proposition can be reasonably considered to be one of a number of similar propositions about which we “know” something (such as that a certain percentage of them are true) which allows us to assign a “probability” to each of them in some plausible way. In other cases there seems to be no reasonable or meaningful way to do this, so it is unclear what it could mean to say that the “probability” that such a proposition is true is such-and-such.

However, it would be absurd to say that, if there is no meaningful way to assign a probability to proposition H (i.e., the expression P(H) is meaningless), it is necessarily meaningless to talk about “evidence” for or against H. If we were to accept this and if I’m right that there are lots of such propositions (including all of those that express possible “natural laws&#8221 , we will be forced to abandon empiricism - and common sense, for that matter - entirely.

So how can we handle such situations? To get some guidance here, consider a case where it is meaningful to assign a probability to a proposition H, and let F be a (possible) observation that might support H. It can easily be shown that P(F|H) > P(F|~H) if and only if P(H|F) > P(H). Thus if F is more likely given that H is true than it is given that H is false, then F makes H more probable, and thus can be considered to be evidence for H.

It seems reasonable to generalize this to say that regardless of whether P(H) is meaningful, F is evidence for H if P(F|H) > P(F|~H). This is done all the time by scientists. Thus, although it’s not at all clear how one would assign a meaningful a priori probability to the hypothesis that the half-life of Radon-222 is less than a week, we can test it by simply starting with a certain quantity of Radon-222 and waiting a week. If more than half of it has decayed, this is evidence for the hypothesis, since the probability that this would happen is much greater if the hypothesis is true than if it’s false.

[By the way, in expressions like P(F|H) the symbol between F and H is a bar, not a slash; it does not indicate division, but means (roughly) “given that”. It’s nonsensical to write P(F|1) > P(F|0), all right, but not because division by zero is undefined. It’s nonsensical because “the probability of F given that 1” doesn’t mean anything.]

Finally, I did not “mix up the hypothesis and the outcome/evidence” in my example. Perhaps you got that impression from statements like “each F(i) is evidence that H is true, because P(F(i)|H) > P(F(i)|~H)”. In view of the explanation above it should be clear that this is exactly what I meant. This is the correct criterion for whether F(i) is evidence for H; it makes sense regardless of whether P(H) can be defined.

4. Miscellaneous issues

That wouldn’t be a fallacy of reification, but in any case it wasn’t what I was doing. My point was that each flip constituted evidence even though it shifted the probability of the hypothesis only a bit. “Proof”, in the sense of proof beyond a reasonable doubt, is often obtained by putting together a lot of pieces of evidence, none of which is all that significant in itself; that doesn’t mean that each piece is not evidence.

It would indeed be a fallacy to argue that since each individual’s belief is evidence, the cumulative weight of near-unanimous belief is very strong. The problem here is not “reification”, but the fact that the beliefs of different individuals are often strongly correlated, so they can’t be treated as independent pieces of evidence. My claim is merely that the fact that a proposition is believed by almost everyone is one piece of evidence for that proposition, it must be considered evidence because it supports the proposition, even though it may be far from conclusive.

The conclusion doesn’t follow from the premise. For example, if (mirabile dictu) I observe that my stockbroker is right 95% of the time when he believes that a stock will double in a year, and he predicts that stock X will double in the next year, It is not a “totally arbitrary and subjective choice” to believe that stock X will double in the next year, even though he has been wrong 50 times out of the last 1000 such predictions.

5. More on belief and professed belief as evidence

If this were so, eyewitness testimony would be completely worthless as evidence.

Suppose that Jones testifies that Smith ran a red light just before he ran into Brown’s car. There are two ways that Jones’s testimony could be false: (i) He could be mistaken; (ii) He could be lying. Thus, in order for his testimony to qualify as “evidence” two things must be true: (i) His profession of belief that Smith ran a red light is evidence that he believes that Smith ran a red light; (ii) His belief that Smith ran a red light is evidence that Smith ran a red light. If belief were not evidence of the truth or falsity of a proposition or if professed belief were not evidence of belief, all eyewitness testimony should be disallowed.

6. What else constitutes evidence?

You say that in the case of the existence of God “there is no objective physical basis on which [this] belief is held.” But aside from the fact that this claim is contested by many theists, you seem to be ruling out a priori a number of other possible types of evidence. For example, if there were a sound, valid logical argument showing that God must exist, this would be “evidence” by any reasonable definition. Also, it’s conceivable that a good argument along the lines of the “fine tuning” argument could be made – that is, an argument that the universe is constituted in such a way that the most plausible explanation is that it was created by an omnimax deity. Finally, it’s possible that, if God exists, a given person could have received a personal revelation from Him. If this could be true of any one person, it could be true for almost everyone. This could be conclusive evidence (for the recipient(s) of the revelation), yet it would not be an “objective physical basis” for belief.

Moreover, there are many beliefs with no “objective physical basis”, which are nevertheless justified. Here are some examples. (1) Rainbows. Today we have a pretty good idea of what rainbows “are” – i.e., what physical conditions result in our seeing them. But for many millennia people knew that rainbows existed simply because they saw them. Long before the science of optics was developed, everyone but very young children realized that rainbows are not physical objects, but that’s not to say that they don’t exist. (2) Emotions. Even today no one has more than the vaguest idea what the physical basis is for emotions. There is not the slightest “objective physical basis” for believing that they exist. Yet almost everyone believes (or rather knows) that they do exist. (3) Mathematical induction. Everyone with an IQ above 90 believes that if a proposition is true for the number 1, and has the property that if it is true for n, it is true for [/i] n+1[/i], then it is true for all natural numbers. Yet there is no “objective physical basis” for this belief, nor can it be proven.

So the argument that a belief is unjustified if it has no “objective physical basis” is incorrect.

7. Postscript

The evidence of near-universal belief, like almost any other evidence, can be “defeated” by other evidence. One such “defeater” would be evidence that the belief is based on emotion or other irrational factors. Another would be that there is no evidence to support the belief other than the widespread existence of the belief itself. One way of putting this is that belief in a proposition is “second-order” evidence for it. Ultimately there must be “first-order” evidence for a belief in order for anyone to be justified in holding it. But a given person need not be in possession of this “first-order” evidence in order for his belief to be justified. The job of the skeptic is to examine whether there is any such “first-order” evidence, and if so whether it is sufficient to justify rational belief. But not everyone need be a skeptic; it is not irrational to refuse to take on this job. And no one can take it on for all possible questions. We all believe a great many things simply because we observe that most other people believe them. Not only is this “evidence”; it is often good enough evidence to justify belief. If not, all of us believe a great many things irrationally – i.e., without any evidence, or at any rate without enough to warrant belief.

Recall that my original point was that the argument that belief in X is justified because, among other reasons, it is widely believed, cannot be dismissed casually on the grounds that the existence of such a widespread belief is not “evidence”. This was basically a call for clear thinking on this point. Since widespread belief is evidence, on what grounds do we (as skeptics) dismiss it when examining whether some proposition is true? I gave my answer in my first post and again above. It has the merit of applying to all beliefs. If anyone has a better one, I’d be glad to hear it.

[ January 17, 2002: Message edited by: bd-from-kg ]</p>

bd-from-kg

theophilus:

Although most of your comments here have nothing to do with the original topic, they’re interesting in a weird, perverse sort of way.

But I’m having trouble following the logic of your argument. It seems that you are making several different claims. I’ll focus on two of them.

1. First, you seem to be saying that non-theists have no way of obtaining any knowledge of any kind; in fact they are unable even to know what constitutes evidence for any proposition. At any rate, this is true for the proposition that God exists:

[Note: Later you claimed that what you were really saying was:

Sorry, no sale. Actually you were saying that non-theists have no basis for determining what would be meaningful evidence for or against the existence of a supernatural being.]

This is echoed by your later statement:

But if all this is true, it is difficult to see how a non-theist can ever escape from his pitiful condition. In particular it would seem to be impossible for him to rationally conclude that God exists. You say:

But how exactly does this work? You said earlier:

So “creation” (i.e., the natural world) cannot provide any “proof” of God’s existence. Moreover, since non-theists have no basis for determining what would be meaningful evidence for or against the existence of a supernatural being, even if such proof were possible and we observed it, we would have no way of knowing that it was even “meaningful evidence” – much less proof – of His existence. So how do we “know” God through the “witness” of His “creation”?

2. Your second claim seems to be that Christians are in better shape in this respect; they can have real knowledge and determine what constitutes meaningful evidence. but your account of how being a Christian gives you access to real knowledge seems to be logically incoherent.

For example:

Now when you were born you didn’t know about these “objective definitions”. How did you come to know about them? Presumably through one or more experiences of some sort. And since you’re human, any experience you had must necessarily have been a human experience. But this means that you were unable to know or understand this experience, since you did not yet have the requisite definitions. A Catch-22.

Next, consider your claim that the Bible gives you a foundation for real knowledge:

OK. But what is the foundation for your belief in the Bible? So far as I can see, nothing at all.

Quite true. So long as you refrain from relying on the Bible to show that the Bible is authoritative, you are not reasoning in a circle. Rather, you are reasoning in a vacuum. If you can give no reason whatsoever for believing that the Bible is God’s word, your “solution” to the unbeliever’s supposed epistemological dilemma is to arbitrarily choose to take the truth of the Bible as a metaphysical axiom.

Well, yes. But if one just arbitrarily chooses any old foundation, one is making a mockery of terms like “truth claim” and “foundation”.

The basis for your claim to “real knowledge” based on the Bible seems to boil down to this: If one assumes that God exists, and that God is completely honest, and that the Bible is the Word of God, one then has “real knowledge” that what the Bible says is true. Well, sure. Similarly, if I assume that Marco Polo existed and that everything he said is true, and that he was the author of Travels of Marco Polo, then I have “real knowledge” about all sorts of things about China. But sane people would not call this “real” knowledge unless I can provide some reasonably persuasive reasons for believing that my assumptions are true. In fact, if I were unable to provide any such reasons but persisted in believing that I really “knew” all the things in Travels I would be headed for the booby hatch. Why shouldn’t your belief that you have “real” knowledge based on the Bible be treated the same way?

And if you reply that, unlike God, Marco Polo is not alleged to be omniscient and all-benevolent, I can simply add the assumption that Polo was inspired by God. Can you prove that he wasn’t? If not, why is my belief in the authority of Travels of Marco Polo less rational than your belief in the authority of the Bible?

Next you claim that it is at least as reasonable to accept the authority of the Bible as to accept the authority of reason:

That’s exactly backward. If the Bible is to be the criteria by which reason is to be judged, then you must demonstrate that your “revelation” is superior to reason.

“Reason” is self-evidently a valid mode of distinguishing truth from falsehood. To reject reason as an “authority” is to embrace insanity. On the other hand, it is hardly self-evident that the Bible is true. So while there is no need to justify using reason as one’s “authority”, there is a need to justify using the Bible as your authority.

Let’s see what happens if we take seriously the notion that the authority of reason requires some “justification”. (This seems to be the heart and soul of presuppositionalism, so it is worth analyzing in some detail.) Your statement of this claim is as good as any:

How does this work exactly? Let’s consider a simple logical argument – call it Arg1:

P1: A
P2: A implies B
C: B

Suppose that I accept P1 and P2. According to you, before I can know that I “ought” to accept C I need a “standard” saying that I “ought” to be logical. OK, let’s add the premise that I “ought” to be logical. Now we have Arg2:

P1: A
P2: A implies B
P3: I ought to be logical.
C: B

It’s not clear how this adds anything to the argument. I still need something to get from “I ought to be logical” to C.

So let’s try changing P3 to something more specific, yielding Arg3:

P1: A
P2: A implies B
P3: I ought to accept C
C: B

But it’s still unclear how P3 adds anything to the argument. Besides, how could I have arrived at it in the first place? Presumably by means of something like Arg4:

P1': I ought to accept the conclusion of a valid argument
P2': Arg1 is valid
P3': C is the conclusion of Arg1
C': I ought to accept C.

OK, but why should I accept the conclusion of this argument? Why, because it’s the conclusion of a valid argument! This is actually yet another argument, which can be formalized in turn, leading to the same question, and so on ad infinitum.

Besides, how did I arrive at P1'? Presumably because I know about the divinely imposed standard that I “ought” to be logical. But how do I know about this standard? There seem to be two possibilities: (i) I deduced it from what I know about God’s nature, or (ii) God “told me so” in some fashion. In the second case, my “knowledge” of P1' is based on an argument that goes something like this:

PP1: God said that I ought to accept the conclusion of a valid argument.
PP2: God never lies.
C1: When God said that I ought to accept the conclusion of a valid argument, He was speaking truly.
P1': I ought to accept the conclusion of a valid argument.

In either case (whether I deduced it or God told me), we can refer to the argument leading to P1' as Arg5. And once again the question arises of why I “ought” to accept the conclusion of this argument. The conclusion itself cannot be used to justify this; it only comes into play after we have accepted the argument. Using P1' to justify the acceptance of Arg5 (and thus of P1' itself) is transparently circular.

By now it should be clear what the problem is. Logic saturates all of our thinking. Without logic we cannot reach the most elementary conclusions. The plain reality is that we must simply take it as axiomatic, without any argument or evidence whatsoever, that we must accept the conclusion of a valid argument, or more generally that we “ought” to be logical.

This is called “sanity”. If you don’t have it in the first place, you aren’t going to acquire it by being told by God that you “ought” to be logical.

The bottom line is that we’re all in the same boat. No one can be sure that he isn’t insane, so no one can be sure that he “really” knows anything, or even that he has any rational basis for what he believes. Assuming that God exists doesn’t help. We know that there are people who are completely insane, yet are absolutely convinced that they are completely sane. There’s no way to now that you aren’t one of those people.

Indeed, I would go further: presuppositionalism (at least the version of it that you present here) is irrational to the point of insanity. And lest you take offense, I might remind you that you have already accused me of insanity and worse. For you said:

Thus you’re directly accusing me of bad faith and low motives. According to you, I know God but refuse to “believe” in Him because of my “willful rebellion”. But if I know God, I know that refusing to “believe” in Him will lead to unending, agonizing torment, whereas believing in what I already know to be true anyway will lead to eternal bliss. Knowingly choosing falsehood over truth, and eternal torment over eternal bliss, is obviously insane.

So thanks very much for your honest opinion. But don’t be surprised if some unbelievers respond to such opinions with hostility and contempt.

theophilus

I apologize for the lack of clarity in my posts. I have no formal training in logic or philosophy, just a lot of reading. So, I'm sure there are more technically accurrate ways to express my arguments.
Without responding to each of your questions/points, let me try to clarify the issue as a whole.
I do not deny that atheists have or can obtain knowledge, on a practical basis. I deny that they can as atheists explain how such knowledge is possible (matter provides no knowledge) or that they can have any confidence that their knowledge is "true."
All our knowledge of God comes as his self-disclosure, i.e,, no one can discover or find God. This includes his revelation through creation. This knowledge is rational - it accords with reason -- but it is not the "result" of rational processes. Thus, no one needs to "study" creation to come to the conclusion that there is a creator. That information is communicated by the creation itself - it "witnesses to" it's creator.
As to your not "believing" this witness, the Bible argues (not I) that all men receive this witness but supress it, i.e., refuse to acknowledge the creator, because of their inherent rebellion agains their creator. This condition is the result of man's sinful nature which he inhereted from Adam, but the denial of God is a willful, personal choice.
There is some confusion over the use of the term "believe." In this case, the term acknowledge is better. In speaking of God, the term belief always implies faith - trust, confidence, submission, i.e., belief "in" God.
I hope this helps clear up some of the misunderstandings.

theophilus

I don't understand your objection. In your earlier message you said:
The fire burns you the first time. "Reason" cannot tell you that it will do so again. There must be a presupposition, not based on experience (we have not experienced the future), that fire will always burn you. Reason goes thus:
1. The fire burned me the first time - experience.
2. Fire always behaves the same way presupposition, not experience.
3. The fire will burn me if I touch it again.

So, our reasoning is based on presuppositions which cannot be "proven" to be true by experience. As a Christian, my presupposition is that the Bible is the word of God and is true and gives me confidence that the world does behave in an orderly, predictible way because that is the nature of its creator.
As an atheist (I'm assuming), how do you have such confidence?

bd-from-kg

theophilus:

In your first post on this thread you asked:

Well, here's an example. It's from the article <a href="http://www.leaderu.com/truth/1truth11.html" target="_blank">Why the Burden of Proof is on the Atheist</a> by Prof. Ralph McInerny.

This seems to be about as pure an example of this kind of argument as one can imagine. Dr. McInerny is a highly respected Catholic scholar, so it is reasonable to suppose that this argument is considered respectable in certain academic circles.

[ January 17, 2002: Message edited by: bd-from-kg ]</p>

theophilus

I'm sorry, you seem to have missed the point of my argument. The question is not whether or not you "ought" to be logical, i.e., accept "c" (a moral statement - and yes, I do challenge even this from an atheistic perspective). The point was, you have no basis for being confident that either "a" or "b" is a meaningful statement about reality.
A. "All men are mortal." This is an inductive statement and is not supported by data. First, no one has "experienced" all men. Second, no one has experienced the future to know whether men born after "now" will continue to be mortal.
B. "Socrates" is a man." Again, an unsupportable inductive argument. Socrates may "appear" to be a man, but since he has never existed before and since no one else can "know" him in his essence, this is a mere assumption.
C. "Therefore, Socrates is mortal." This statement is not only invalid, it is worse - it is useless. You might as well say, "blah, blah."

In order to be valid, the propositions would have to be so restricted and qualified that they would ultimately be meaningless.
A. "All men who have existed up to this point and who have been observed (passing over the question of the reliability of sense perceptions) to have died at some point and are known not to have come back to life at a later point are mortal," or put more simply, "all men who are mortal are mortal."
B. I'll simplify this: "Socrates may be one of the men who are mortal."
C. "Therefore, Socrates may be mortal."
You'll be bound to admit that this statement doesn't contribute a great deal to the store of knowledge.

If you like, we can discuss the impossibility of establishing a moral necessity for being logical
on an atheist foundation.

theophilus

Sorry, no cigar. Jamiel's original statement was,
"I've read some interesting claims on this board by theists relating to the fact that so many human beings believe in some form of god/religion. The posters say that this is supportive of the existence of god(s) but that it is not simply an appeal to numbers. I've seen two forms of this argument:"

Unless you found this quote "on this board," my question still stands.

Thanks.

Theli

About confidence...
I'm quite aware that it's possible that a natural disaster can wipe me out, just like that...
Or that someone can just suddenly run me over when I'm walking in the street. Unpredictable things happen everyday. To "bad" aswell as "good" people. I have no illusions of being shielded from the world by some allpowerful being.
My only confidence is credibility. The chance that I will die from a natural disasted today is MUCH smaller than the chance that I won't.

Dirk the Daring

Exactly. There is a different standard of admissibility in a rigorous or formal mode than in a "common-sense" one. Philosophical and scientific discussions are typically conducted in a formal mode. So in the context of a philosophical forum like this one, while it might be occasionally pertinent to point out the discrepancy in modes, one should never confuse the two by bringing in analogies like jury trials.

In the formal mode, the number of people who believe X has no bearing at all on the truth of X. An awful lot of fallacies are propagated by people who refuse to, in the words of the old Chinese Zen masters, decide for themselves if the water is hot or cold.

Photocrat

Eh?

Saying that "because lots of people believe, there must be a god" is ad populum, a logical fallacy.

Saying "X is not extrodinary because lots of people believe it" is logical, given the definition of extrodinary, though it doesn't really prove much.

Saying "X having the property of Y is more likely under hypothesis A than under hypothesis B, X has property Y, therefore this is evidence for hypothesis A" is also logical. *

#1 is illogical proving nothing, #2 doesn't even purport to be evidence & #3 is highly dependant on the data & hypotheses, as to how strongly the hypothesis is confirmed.

However, none of these can really prove anything, save perhaps #3 and that only if the degree to which it confirms A is far greater than that to which it confirms B.

[* I hope I phrased this right, but I'm tired. Look up Bayes theorem.]