Bricker is a disingenous punk.

[begbert2]

[QUOTE=Hentor the Barbarian]
We have gotten no further with this than before. The original question was whether one of the claims of two people could be treated as more credible than another. To do that, you must have evidence that would describe their credibility.

If you say that their credibility is irrelevant, then you are avoiding, altering or just ignoring the original question. It makes no logical sense to say, “You two are equally credible, therefore B, you are less credible than A.”

I fully understand that, ignoring the credibility of two people altogether, you would be best served if you had to choose whether one event happened at all over another, you may as well go with the probability that the event happened at all. This is as if you had drawn the two individuals randomly from the population, or if you felt that people were perfectly incredible. But then you cannot state that one person is more credible than another. This would be true if two people hadn’t walked into your office in the first place.

So you have two specific individuals making specific claims. You sit there, and you say to them, “The probability of X happening to the average person is greater than the probability of Y happening to the average person.”

They look back at you, stunned, and say “What the fuck does that mean? I’m not the average person! What about the two of us?”

So, you say, “All I can say is: The probability of X happening to the average person is greater than the probability of Y happening to the average person.”

They say, “But how can we be the average person? The average person doesn’t make a claim of rape! Doesn’t that make it completely inappropriate to apply statistics for the average person to us, since you know we are necessarily different than the average person?”

You respond, “I’m ignoring you altogether, since I have no further information to go on, and, although I’ve already concluded you are necessarily equally credible, for some reason I feel compelled to disregard this and apply general population statistics to you. For some reason, I feel compelled to see A as more credible.”

B: “But didn’t you just say that we were equally credible?”

You: “Yes. But I’m ignoring that.”

B: “What the fuck is wrong with you? Do you have a problem with B people?”

You: “Oh, no no no. I’m being perfectly logical. Don’t get all emotional!”

A: “Hey, look over there. That man just won the lottery!”

You: “No, he didn’t. Things that don’t happen more often than not just never happen.”
[/QUOTE]
Firstly, I have been concerning myself with wether a particular claim is credible, not wether a particular person is credible. If you don’t see a difference there the only way I can think to explain it is that you can find a claim to be false, and still find the person to be credible.

Now, when I assume that two people are equally credible, and that they are not (neither of them) 100% credible, then what does that tell me? It tells me that they might lie, or be mistaken, or whatever. Either of them. There is also no reason whatsoever for me to assume that wether one of them lies in one particular instance has any effect on wether the other of them is lying at the same time. Correct?

So. We have two people who are each equally credible, and can lie; however, merely looking at the person doesn’t tell me wether they are lying. This means it’s not like “You have scuzzy hair and a bad necktie, so I doubt your credibility,” or like “You’re a whitey and I’m a racist, so I doubt your credibility.” Each person in the scenario is intrinsically trustable, based on their own personal attributes. Like identical twins; I can’t tell them apart even.

And then one of them tells me that he likes the taste of chicken, and the other one tells me he’s actually President Abraham Lincoln, brought to life again my the dark magic of Ra, as used by a japanese ex-wiccan with a drug problem…

It’s the claim itself that is less credible, not the person.

Secondly, I would like to point out that finding a claim to be incredible is not equal to automatically assuming it’s false. It’s just unlikely. If you’re going to put words in my mouth, the words would more likely be “What, he won the lottery? Him right there? Are you kidding?”, not your work of fiction.

And thirdly, as an aside, I would also like to point out that any scenario that has me actually conversing with the claimants is moving the goalposts. I would get a lot more information from them than their gender, race, claim, and name if we do that.

[begbert2]

Sorry; I meant “Each person in the scenario is intrinsically equally trustable, based on their own personal attributes”. My bad; sorry about that. (If they were intrinsically trustable, it would be a short debate.)

[RaftPeople]

[QUOTE=begbert2]
False. Using what information you have available is better than randomly choosing.
[/quote]

Using what information you have available is better than randomly choosing.

Unfortunately you have ZERO information regarding how the frequency of an event influences the accuracy of the claim.

The other information you do have, which is the relative difference in frequency of events does not tell you ANYTHING about the chances the claim is a false claim.

Duh.

FYI: I know we are in the pit, but my philosophy is that, when debating, it’s most productive not to insult, etc.

And you’re speculating that the results of studies that we don’t have might indicate a fact that would result in a different conclusion than the known facts. That’s nice. We don’t have such studies, though, do we? For all we know, they might support the conclusions the known data indicates. They might cement the issue beyond all doubt, 100%. They might say nothing. We just don’t know. Because we don’t have them.

If we had such studies, we’d use them and be glad of it. Heck, if we had some oracle that told us with perfect accuracy wether a given claim was true, that would be even better! Should we refuse to draw a conclusion until we have have that, too?

No, we can draw conclusions, let’s just make sure they are accurate. For example, here is a conclusion we could logically draw “with the supplied information we are unable to say which claim is more likely to be accurate.”

Now, common sense tells us that the rarer the event, the more implausible claims that it occurred are.

No, common sense does not tell us that.

This is exactly the problem.

Sometimes common sense is great, sometimes common sense returns the exact opposite result.

For example, when questioned whether a bowling ball or a marble will fall faster, I personally know a substantial number of people that have answered bowling ball. As a matter of fact, I myself did the first time I was asked the question. This is just one example, I could go on for a long time with examples of common sense failing people.

[Hentor_the_Barbarian]

Firstly, I have been concerning myself with wether a particular claim is credible, not wether a particular person is credible. If you don’t see a difference there the only way I can think to explain it is that you can find a claim to be false, and still find the person to be credible.

All we have here is people making claims. It’s a person’s claim. That’s it. A claim cannot exist without a person to make it. If a claim happened in the woods with nobody there to make it, would it make a sound? An event may happen independently of a person, and rapes may happen independently of claims. But trying to distinguish between the credibility of a person and their claim when all you have is a hypothetical person’s claim is ludicrous.

It would, nevertheless, be interesting to hear how you would find a claim to be false but a person making that claim to be credible.
[QUOTE=begbert2]
And thirdly, as an aside, I would also like to point out that any scenario that has me actually conversing with the claimants is moving the goalposts. I would get a lot more information from them than their gender, race, claim, and name if we do that.
[/QUOTE]
How is that “moving the goalposts”? I’m not sure you know what that phrase means. Moving the goalposts means changing the point you are asking your debate opponent to prove or the criteria by which he might prove it. I’m not asking you to demonstrate anything other than how you can state that two people making claims are equally credible but one is more credible than the other.

Do you agree or disagree that once they make a claim of having been raped, they are no longer “the average person” to whom general population probabilities may be applied?

Nevertheless, the issue has always been “two women claim rape.” Yes, for comedic and illustrative purposes, I had them engage in a discussion about how one might dismiss their credibility as claimants.

As to the rest, it is non-responsive to the matter.

[begbert2]

[QUOTE=RaftPeople]
The other information you do have, which is the relative difference in frequency of events does not tell you ANYTHING about the chances the claim is a false claim.
[/QUOTE]
You are nearly correct; it does not tell you much about the chances the claim is a false claim. Unless the relative difference is staggeringly huge, of course.

The relative probabilities of the odds of the event(s) occuring tell us something if the difference is staggeringly huge; I think we can agree on that. Zombie Lincoln and all. Is there a magic point of probability after which they don’t tell us anything at all? Or does the persuasive impact of the odds merely reduce as the long odds get smaller, until it is easily outweighed by other factors; only to reach zero persuasion when the relative odds are equal? I like the second explanation better; it doesn’t require indefinable ‘magic points’.
[QUOTE=RaftPeople]
FYI: I know we are in the pit, but my philosophy is that, when debating, it’s most productive not to insult, etc.
[/QUOTE]
‘Duh’ is an insult? Well, you know what they say: insult the post, not the poster; insult the argument, not the arguer. I find the assertion that the content of the claim has no bearing on the claim’s credibility to be absurd on the face of it; there are a variety of ways I can convey that belief. This time, I chose three letters.
[QUOTE=RaftPeople]
No, we can draw conclusions, let’s just make sure they are accurate. For example, here is a conclusion we could logically draw “with the supplied information we are unable to say which claim is more likely to be accurate.”
[/QUOTE]
We can never make certain of of the accuracy of a claim made on statistics alone. Not even if those statistics happen to be about the accuracy of claim rates. The best you’ll get is a matter of reducing that plus-minus margin of certainty on your result; and yet again, the difference between having the claim stats or merely population stats being one of scale, and not anything fundmental.

Fortunately, certainty is not a requirement of the problem. Merely a best guess. Based on available facts.

And I await your convinicing logical proof that the content of a claim has no bearing on its credibility. You’ll have to show me, because without some kind of support, I find that claim to be incredible (regardless of who happens to be making it).
[QUOTE=RaftPeople]
No, common sense does not tell us that.

This is exactly the problem.

Sometimes common sense is great, sometimes common sense returns the exact opposite result.

For example, when questioned whether a bowling ball or a marble will fall faster, I personally know a substantial number of people that have answered bowling ball. As a matter of fact, I myself did the first time I was asked the question. This is just one example, I could go on for a long time with examples of common sense failing people.
[/QUOTE]
Wait, you tell me that common sense does not support my assertion, and then launch into a full-bore attack on the concept of common sense itself? Why would you do that if you thought common sense didn’t support my position? Are you being truly honest with yourself here?

Regardless; since we don’t actually know anything about the race/claim rate ratios, they don’t even enter the equation. The only place ‘claim’ occurs in the problem is in the mere fact that there are claims being made; there is nothing else in the problem to help us extract any more information from the fact claims are made, so we don’t. (You’ll note that common sense is not cited in the argument as a reason for accepting the odds; I mentioned it as a reason not to give credence to the strawman about studies that don’t, in fact, exist. Not that one actually needs a counterargument to topple strawmen, of course.)

[begbert2]

[QUOTE=Hentor the Barbarian]
All we have here is people making claims. It’s a person’s claim. That’s it. A claim cannot exist without a person to make it. If a claim happened in the woods with nobody there to make it, would it make a sound? An event may happen independently of a person, and rapes may happen independently of claims. But trying to distinguish between the credibility of a person and their claim when all you have is a hypothetical person’s claim is ludicrous.

It would, nevertheless, be interesting to hear how you would find a claim to be false but a person making that claim to be credible.
[/QUOTE]
So, did your mommy ever tell you anything about a guy named “Santa Claus”?
[QUOTE=Hentor the Barbarian]
How is that “moving the goalposts”? I’m not sure you know what that phrase means. Moving the goalposts means changing the point you are asking your debate opponent to prove or the criteria by which he might prove it. I’m not asking you to demonstrate anything other than how you can state that two people making claims are equally credible but one is more credible than the other.
[/QUOTE]
In the original argument, all I know is their gender, race, the claim they made, and (as a late-game addition) their names.
[QUOTE=Me]
And thirdly, as an aside, I would also like to point out that any scenario that has me actually conversing with the claimants is moving the goalposts. I would get a lot more information from them than their gender, race, claim, and name if we do that.
[/QUOTE]
'Nuff said.
[QUOTE=Hentor the Barbarian]
Do you agree or disagree that once they make a claim of having been raped, they are no longer “the average person” to whom general population probabilities may be applied?
[/QUOTE]
What, have they stopped being part of the total population now? Knowing that they claimed to be raped is an additional nugget of information, which would be potentially really useful if we, you know, had any way to use it. Failing that, they’re still people…and we have some odds about what happens to people.

[Hentor_the_Barbarian]

[QUOTE=begbert2]
So, did your mommy ever tell you anything about a guy named “Santa Claus”?
[/quote]
That is a failure to demonstrate your point. We are talking about one person and one claim (or two people and two claims, but you understand the point). You are now trying to make a statement about the overall credibility of a person over a lifetime of claims. Tell me how you would find a claim to be false but a person credible in the context of one claim from one person, with no other information.

What, have they stopped being part of the total population now? Knowing that they claimed to be raped is an additional nugget of information, which would be potentially really useful if we, you know, had any way to use it. Failing that, they’re still people…and we have some odds about what happens to people.

Yes, they stopped being like the average person from the population the moment they made the claim of rape, since that is not true of the average person in the population. It’s appropriate to apply the proper statistics for that subpopulation of the larger population in the same way that you want to apply the proper statistics from the subpopulations of African American and Caucasian women.

Or do you not want to do that, now? If not, tell me how you determine the relative credibility using the statistic of how many times all women in the whole population are raped. Want to use subpopulation statistics sometimes but not others? Why is that?

Just because you don’t know what the appropriate statistic is does not make it reasonable to base a decision about credibility on inappropriate statistics.

[Contrapuntal]

All I want to know is this – is Bricker a disingenuous punk or not? Have either of the elements A) disingenuousity, or B) punkitude, been proven? Because I cannot get my head around all these statistics.

It does seem to me, though, that if the alleged victim (AV) was lying, it has nothing to do with her being black. It could just be that she is a liar.

[monstro]

[QUOTE=Contrapuntal]
All I want to know is this – is Bricker a disingenuous punk or not? Have either of the elements A) disingenuousity, or B) punkitude, been proven?
[/quote]

Bricker is someone who refuses to take complete responsibility for things that he has said. I attempted to get him to do so, but he would rather protray himself as a victim of one great big misunderstanding. His performance is a tiring one to watch, so I’ve given up on him.

Because I cannot get my head around all these statistics.

The statistics are a red herring. The question is real simple: If two people report a being raped by white male and you know nothing about them except their race, which of the claims is more credible? Some would have you believe that the white person is more credible because white women are more often raped by white men. Others (myself included) say that both people are equally credible in the absence of evidence because the claims are equally plausible.

It does seem to me, though, that if the alleged victim (AV) was lying, it has nothing to do with her being black. It could just be that she is a liar.

We aren’t talking about the Duke accuser. We are talking about a certain kind the “logic” which leads people to find claims of white-on-black rape less than fully credible.

[you_with_the_face]

Uh, the previous post was from me, not monstro. Mea culpa.

[wring]

actually, the original question was much simplier. when a black woman accuses several white men of gang rape, is it appropriate to consider one years’ data on the relative frequency of white male on black female rape to disbelieve her claim (including the fact that the same years data indicated that when gang rape occurred, it was more likely to have been committed by white males - w/o reference to race of victim).

[Contrapuntal]

[QUOTE=you with the face]
Uh, the previous post was from me, not monstro. Mea culpa.
[/QUOTE]
Sorry** face**. I was using my *so rarified that it is * *almost effervescent *ironic wit to slyly comment on how long this thread has gone on, and how much it resembles its predecessor, in that it has gone off the rails into a discussion about statistics. I was being a bit of a smartass.
For the record, and speaking from a laymans point of view with respect to statistics, I cannot see how the races of the principles involved would have any bearing at all on how credible they are.

[begbert2]

[QUOTE=Hentor the Barbarian]
That is a failure to demonstrate your point. We are talking about one person and one claim (or two people and two claims, but you understand the point). You are now trying to make a statement about the overall credibility of a person over a lifetime of claims. Tell me how you would find a claim to be false but a person credible in the context of one claim from one person, with no other information.
[/QUOTE]
Lord, what a weasel-fest. How in the hell am I supposed to demonstrate a case where the person’s credibility and the claim’s credibility differ (which is necessary to, you know, demonstrate that they differ, which is the entire point I was making) if you restrict me from bringing information allowing us to determine things about the example person’s credibility into it? Sorry sucker, I’m not falling for it.

You have several options:

1. assert that, since Santa Claus is highly unlikely to exist, that one should never believe anything your mother says.

2. assert that, since your mother is at least moderately credible, one should seriously entertain that there’s an ancient fat man that exceeds the speed of light every christmas in a reindeer-propelled flying sled.

3. concede that, yes, generally honest people are capable of saying things that are obviously not true, and that generally unreliable people are able to make highly credible statements now and then (“Hey, that hurts!”); and therefore you concede that a person’s general credibility is not the same thing as any/all of their claims’ credibility.

4. Weasel wildly like a disengeneous punk.

5. Ignore what I’ve said and/or pretend I said something else.

I sincerely hope to be surpised by your reply.
[QUOTE=Hentor the Barbarian]
Yes, they stopped being like the average person from the population the moment they made the claim of rape, since that is not true of the average person in the population. It’s appropriate to apply the proper statistics for that subpopulation of the larger population in the same way that you want to apply the proper statistics from the subpopulations of African American and Caucasian women.

Or do you not want to do that, now? If not, tell me how you determine the relative credibility using the statistic of how many times all women in the whole population are raped. Want to use subpopulation statistics sometimes but not others? Why is that?

Just because you don’t know what the appropriate statistic is does not make it reasonable to base a decision about credibility on inappropriate statistics.
[/QUOTE]
Actually, we’re given the statistic that wow rape is more common than wob rape. And that’s exactly as far as I take the subpopulation angle. Your attempts to imply otherwise are duly noted and reviled; your ridiculous conclusion that has nothing to do with the statements preceding it is similarly dismissed.

You are welcome to try to support your assertion that Axiom C is and ‘inappropriate’ statistic. Please confine yourself to facts that are not unavailable in the argument at hand.
On preview:

[QUOTE=you with the face]
Some would have you believe that the white person is more credible because white women are more often raped by white men.
[/QUOTE]
But not much more credible, and only if the best statistics you have available support it, and even then we’re not recommending that you alter your behavior in the slightest over the piffling difference in credibility you’re likely to unearth. Or at least, that’s what I’m arguing.

And with the scenario that wring presents, it may be appropriate for an individual person to consider the data, if they were, you know, bored; but unless they turn out to be Zombie-Lincoln odds it’d be damned irresponsible to try to present those odds as something useful in a court or over the media. Heck, non-statisticians might hear about it, and blow it all out of proportion!

[Hentor_the_Barbarian]

[QUOTE=begbert2]
Lord, what a weasel-fest. How in the hell am I supposed to demonstrate a case where the person’s credibility and the claim’s credibility differ (which is necessary to, you know, demonstrate that they differ, which is the entire point I was making) if you restrict me from bringing information allowing us to determine things about the example person’s credibility into it? Sorry sucker, I’m not falling for it.

You have several options:

1. assert that, since Santa Claus is highly unlikely to exist, that one should never believe anything your mother says.

2. assert that, since your mother is at least moderately credible, one should seriously entertain that there’s an ancient fat man that exceeds the speed of light every christmas in a reindeer-propelled flying sled.

3. concede that, yes, generally honest people are capable of saying things that are obviously not true, and that generally unreliable people are able to make highly credible statements now and then (“Hey, that hurts!”); and therefore you concede that a person’s general credibility is not the same thing as any/all of their claims’ credibility.

4. Weasel wildly like a disengeneous punk.

5. Ignore what I’ve said and/or pretend I said something else.
   [/quote]
   Okay, you’re pretty much worthless. I asked you in the context of the hypothetical under discussion to demonstrate how a claim and a person’s credibility could be distinct. You’re the one going on about how the credibility of this person claiming rape, who we know so little about, could be disentangled from her claim.

Your best shot to demonstrate that is to suggest a forced choice between Santa Claus and my mom? And then you get all pissy and call me a weasel when you’re called on your bullshit?

You’ve got nothing, but you’re spinning wildly with it. I’d suggest you go fuck off, but I’d rather you keep talking. That way, you could only make yourself look worse.

[begbert2]

[QUOTE=Hentor the Barbarian]
Okay, you’re pretty much worthless. I asked you in the context of the hypothetical under discussion to demonstrate how a claim and a person’s credibility could be distinct.
[/QUOTE]
What you really asked:
[QUOTE=Hentor the Barbarian, post #564]

It would, nevertheless, be interesting to hear how you would find a claim to be false but a person making that claim to be credible.
[/QUOTE]
Notice how you don’t so much as mention “the context of the hypothetical under discussion”. Just pointing this out.

I notice you chose response number 5. I am not even remotely surprised. I forgot to mention option 6, though: raise a shitstorm of random ad hominem attacks in an attempt to distract attention from how your failure to respond. Frankly, it didn’t even occur to me.

Your flimsy position depends on one or both of the following two statements being true:

1. If two people are assumed to have equal credibility in the absence of other information, then based on that alone it should also be assumed to imply that any claims they make have equal probability, regardless of whatever other information might come to light.

2. The content of a claim has no effect whatsoever on its plausibility.

Because obviously, if we’re allowed to alter our opinions about a claim’s credibility based on other evidence, and if the content of the claim is able to indicate that we should modify the credibility of the claim, then it is definitely, absolutely, certainly, the case that, since the credibility of a claim may be altered by evidence about the claim itself from the assumed default crediblity of the person, then the credibility of a claim and a person’s credibility ARE distinct.

I like statement 1. It’s the one that says that evidence should be forbidden at trials, or at the very least ignored by all present.

I also like statement 2. It’s the one that says there is no such thing as a true or false statements. (Only honest and dishonest people!)

Go ahead, defend one of these positions. I’d love it, for numerous reasons. Or concede the point. I’d be shocked and amazed, and my opinion of you would raise considerably (not that you care about my opinion…).

Or just fling shit. Whatever makes you feel better about yourself.

[RaftPeople]

[QUOTE=begbert2]
The relative probabilities of the odds of the event(s) occuring tell us something if the difference is staggeringly huge; I think we can agree on that.
[/quote]

If we are trying to predict the future, the relative probabilities do tell us something. If we are trying to guess at past events in the absence of any claims, then the relative probabilities do tell us something.

The minute we introduce a 2nd event, that of a human claiming something happened, then the model changes significantly.

The problem immediately shifts to one of accuracy of reporting of events. We are no longer concerned with how often something happens. We are now concerned with how often the person’s claim is accurate. In the absence of that information we assume they are equally accurate.

It is at this point you continue to make your logic error.

You assume that, because both peoples claims are equally credible, we can collapse the problem back to the type of problem where no claim is being made, the type of problem where we are just guessing at events based on probabilities.

I realize you think that “common sense” allows you do this, but I can guarantee you that nowhere in any logic or mathematical proof that I have ever seen was one of the steps labeled “the argument of common sense.”

If you can provide something better than “common sense” I would be happy to retract my 38 posts on this topic, I personally just enjoy seeing the text mapped into a good model.

I guess we will just have to leave it at the following:

1. We agree the credibility of the claim is unknown and thus assumed to be equal (you agreed to this in a previous post)
2. We disagree that “common sense” allows us to assign a different credibility to each claim despite an equal level of credibility

[begbert2]

[QUOTE=RaftPeople]
The problem immediately shifts to one of accuracy of reporting of events. We are no longer concerned with how often something happens. We are now concerned with how often the person’s claim is accurate. In the absence of that information we assume they are equally accurate.
[/QUOTE]
Does this hold even if one of the claims is absolutely ridiculous? If a person tells you that he saw our dear, old friend, Zombie Lincoln walking the street, do you immidiately give it equal credence with “I like cheese”, merely because somebody claimed it?

If not, at what probability do you decide that probabilities are irrelevent?

Unless you can answer this, I will find it difficult to accept they your position is more reasonable than mine.

[DMC]

[QUOTE=begbert2]
Now, common sense tells us that the rarer the event, the more implausible claims that it occurred are.
[/QUOTE]
No it doesn’t.

If one person tells you they are a dental assistant and another tells you they were at Woodstock in '69, which one do you believe?

If I tell you that I subscribe to SQL Server magazine, and that I’m currently SpecOps (Ranger, SEAL, etc.), and if only one is true, which one do you believe?

Who’s more credible, a person who tells you they live in Lawrence, KS, or a person who says they are members in the Screen Actor’s Guild?

[RaftPeople]

[QUOTE=begbert2]
Does this hold even if one of the claims is absolutely ridiculous? If a person tells you that he saw our dear, old friend, Zombie Lincoln walking the street, do you immidiately give it equal credence with “I like cheese”, merely because somebody claimed it?

If not, at what probability do you decide that probabilities are irrelevent?

Unless you can answer this, I will find it difficult to accept they your position is more reasonable than mine.
[/QUOTE]

This holds with claims that have happened.

I do agree that claims regarding events that have never happened require a different approach.

[begbert2]

[QUOTE=DMC]

@begbert2

Now, common sense tells us that the rarer the event, the more implausible claims that it occurred are.

No it doesn’t.

If one person tells you they are a dental assistant and another tells you they were at Woodstock in '69, which one do you believe?

If I tell you that I subscribe to SQL Server magazine, and that I’m currently SpecOps (Ranger, SEAL, etc.), and if only one is true, which one do you believe?

Who’s more credible, a person who tells you they live in Lawrence, KS, or a person who says they are members in the Screen Actor’s Guild?
[/QUOTE]
We’re talking about making best-guesses about events about which we have no other information about than their probabilities of occurrence. Absent other information, we use the odds, if we have them.

Are we certain to be right? Of course not. Even if I knew any of the probabilities for the statements you listed, unless the odds are 100% or 0%, then I have a chance of being wrong. One can’t avoid that when they’re using statistics. What I’m claiming is ‘common sense’ is that if you know the odds, guessing as the odds indicate will, on the average, give you better luck then guessing at random.

[QUOTE=RaftPeople]

[QUOTE=begbert2]

[QUOTE=RaftPeople]
The problem immediately shifts to one of accuracy of reporting of events. We are no longer concerned with how often something happens. We are now concerned with how often the person’s claim is accurate. In the absence of that information we assume they are equally accurate.
[/QUOTE]
Does this hold even if one of the claims is absolutely ridiculous? If a person tells you that he saw our dear, old friend, Zombie Lincoln walking the street, do you immidiately give it equal credence with “I like cheese”, merely because somebody claimed it?

If not, at what probability do you decide that probabilities are irrelevent?

Unless you can answer this, I will find it difficult to accept they your position is more reasonable than mine.
[/QUOTE]

This holds with claims that have happened.

I do agree that claims regarding events that have never happened require a different approach.
[/QUOTE]
What, you don’t like Zombie Lincoln? But he’s a tradition now! I’ve been using him to represent a claim that is highly unlikely but not impossible. (What, you know that it’s impossible to reanimate the dead? Don’t tell the christians!)

If you don’t like that, then, how about “I am Osama Bin Laden.” The odds of randomly selecting a person who is able to say that truthfully are pretty low, I think. Or “I am currently the president of the United States.” Similarly, your random survey will turn up few hits for that. Or on another angle: “A giant meteor is going to hit the earth tomorrow!”? That happened at some point in the past, I believe. It could happen again.

It really doesn’t matter what the claim is; there are plenty of claims that are really, really unlikely when coming from the mouth of random person X that you know nothing else about. My point remains the same; think of the wildest claim you’re willing to accept as, er, ‘already happened’, and tell me truthfully if you consider it equally credible with something truly probable, statistically speaking, when heard from an unknown person. “I have worn socks.” “I have my sight.” Whatever.

I’ll be frank with you; I want you to concede that there are some claims wherin the claim itself has a level of credibility that supercedes the assumed credibility we grant to an unknown speaker. If you concede that, you’re in the unenviable position of either arguing for a ‘magic point’ that divides the probabilities that matter from those that don’t, or simply accepting that it’s not unreasonable to think that all claims have an inherent probability of their own, independent of their unknown source, absent other information.

With many of the more reasonable claims this probability is often sufficiently close to 50/50 that the credibility we generously grant to the speaker dwarfs the claim’s inherent credibility. Because of this, for the claims that are not suspicious or outrageous, we find ourselves acting, very reasonably, as though the only credibility that matters is the speaker’s. This is fine, and gives pretty much everyone what they want with regards to speaker credibility, I think; nonetheless those picky math types are still right in saying the odds of the inherent claim do make a small, tiny, perhaps undetectable difference in the odds of the ‘speaker adjusted’ claim.

Let me know what you think.