Evidence for Jesus

[Mojo]

Do you think it likely that there is a Jewish lawyer named Daniel living in NYC?

Why or why not?

It may be likely, but that isn't actually evidence that it is so.

If I throw a dice, it is likely that I won't throw a six. That doesn't mean I won't throw one.

 

[HansMustermann]

I'm not moving towards an analogy that mirrors Jesus in every way, I'm constructing analogies that clarify how we use evidence.

As for Harry Potter and Superman. Let's stop for the moment talking about figures we KNOW to be false.

Why? If a method is sound, then it can be tested on some cases where you already know the result.

1. If a method somehow only works on cases where you can't verify the results, then it doesn't work, period.

2. If a method can't say "no", or you can't test it on any cases where you know it should say "no", it's worthless, because there is no way for it to falsify anything with it.

If I propose that I make the sun rise, and my criterion is that I live in a real town, therefore the sun rises because of me, you can't test it on only the positive case it's supposed to support. Until you see that it can give a "no" for other people known to NOT be the reincarnation of Ra (if nothing else, because I'm the one, and they can't be too), it's worthless.

 

[Cavemonster]

It may be likely, but that isn't actually evidence that it is so.

The likelihood isn't the evidence, the reasons we come to our viewpoint about the likelihood are the evidence.

The likelihood is the conclusion. It can be mapped on a scale from impossible to certain, and information we receive that gives us good reason to move our position for a claim on that scale is evidence.

If this is a semantic disagreement, and you feel that information is only "evidence" if it creates certainty all by itself, I'm not too interested in continuing.

 

[Cavemonster]

Why? If a method is sound, then it can be tested on some cases where you already know the result.

Why? Because I think I think it takes us on a tangent. A few actually.

I mentioned in my post that you quoted that one of those tangents was the fact that information can be evidence for a position, even if that position turns out to be wrong. That may need to be covered here, but others are definitely unnecessary.

Another tangent is about the order information is received in and its status as evidence based on that. In the case of Harry Potter and Superman, even before we receive the information about location, we learned that the stories were written as fiction.

These and many more are all very interesting topics, but they could each fill up their own thread.

I'd rather concentrate for the moment on how broad, demographic level information can be valid evidence about a specific person. And I think that's what you're interested in too, and we can arrive at a mutual understanding of that better by pursuing a positive discussion of how evidence can work that way rather than the negative discussion of why I think those fictional examples are not comparable.

 

[HansMustermann]

Which of these is the real person?

A) My name is Daniel Stein, I'm a lawyer living in New York City, New York. I'm Jewish. I work at the law firm my father started, live in a very expensive apartment, and I ate bagels with lox from a local deli last Tuesday.

B) My name is Daniel Stein, I'm a lawyer living in Cafombo, Angola. I'm Jewish. I work at the law firm my father started, live in a very expensive apartment, and I ate bagels with lox from a local deli last Tuesday.

How about neither? You haven't supported that the person saying that actually is any of that or has done any of that.

 

[Mojo]

The likelihood isn't the evidence, the reasons we come to our viewpoint about the likelihood are the evidence.

But that evidence is only evidence for the proposition that it is likely that there is a Jewish lawyer called Daniel in New York. It isn't evidence that there actually is one.

 

[Cavemonster]

How about neither? You haven't supported that the person saying that actually is any of that or has done any of that.

I'm giving it to you as a true premise within the hypothetical that one of these is correct.

Someone filled out true information and the machine replaced the actual location with a false location for one of the options.

I'll bet you have a hunch which one is more likely.

 

[Cavemonster]

But that evidence is only evidence for the proposition that it is likely that there is a Jewish lawyer called Daniel in New York. It isn't evidence that there actually is one.

I don't see those as distinct propositions. One is just a level of certainty of the other.

And I don't think you really do either. All evidence just increases the perceived likelihood of a proposition. That's what evidence is.

 

[HansMustermann]

Why? Because I think I think it takes us on a tangent. A few actually.

I mentioned in my post that you quoted that one of those tangents was the fact that information can be evidence for a position, even if that position turns out to be wrong. That may need to be covered here, but others are definitely unnecessary.

Another tangent is about the order information is received in and its status as evidence based on that. In the case of Harry Potter and Superman, even before we receive the information about location, we learned that the stories were written as fiction.

Yes, and that's how we verify those methods. To actually verify anything, yes, you must already know the answer. Arguing that you should be allowed to use something only on stuff where we can't test the result is just as bogus when you do it, as when the mediums and dowsers somehow only work when you're not testing the result.

What you present as a tangent is exactly testing if the method works. And that's not a tangent, but crucial to knowing its accuracy.

These and many more are all very interesting topics, but they could each fill up their own thread.

I'd rather concentrate for the moment on how broad, demographic level information can be valid evidence about a specific person. And I think that's what you're interested in too, and we can arrive at a mutual understanding of that better by pursuing a positive discussion of how evidence can work that way rather than the negative discussion of why I think those fictional examples are not comparable.

Well, yes, and the dowser and homeopaths would also rather concentrate on just the positives. So what?

 

[HansMustermann]

I'm giving it to you as a true premise within the hypothetical that one of these is correct.

Someone filled out true information and the machine replaced the actual location with a false location for one of the options.

I'll bet you have a hunch which one is more likely.

But that's exactly an unsupported assumption: that one of these is correct. In reality I could say A too, but none of that would be actually true. That a city being real supports the actual person who did all that is bogus.

Real life doesn't work by such bogus dichotomies, and knowing that one of the two must be true. Cthulhu involving the fictive city of R'lyeh doesn't make Christianity true because it mentions the real city of Jerusalem.

In reality your A and B are orthogonal. A can still be false even if B is false. They're independent. After checking the city, what you have is that B is disproven by using a false city, but A is still neither supported nor disproven. Not being disproven is not the same as supported.

 

[Mojo]

The likelihood is the conclusion. It can be mapped on a scale from impossible to certain, and information we receive that gives us good reason to move our position for a claim on that scale is evidence.

If this is a semantic disagreement, and you feel that information is only "evidence" if it creates certainty all by itself, I'm not too interested in continuing.

The likelyhood might change the amount or nature of evidence that would be needed to support the existence of the lawyer. For example, given that it is likely that there is a lawyer called Daniel in New York, an entry for someone called Daniel in a directory of New York lawyers would probably be enough for me to accept that there is such a lawyer. This still wouldn't make it absolutely certain, of course; there could be a mistake or misprint, or the lawyer in question could have died or moved to another city since the directory entry was created.

On the other hand, I would want more than a directory entry to support a claim that there is a lawyer called Daniel who can turn water into wine and come back from the dead.

 

[Cavemonster]

Yes, and that's how we verify those methods. To actually verify anything, yes, you must already know the answer. Arguing that you should be allowed to use something only on stuff where we can't test the result is just as bogus when you do it, as when the mediums and dowsers somehow only work when you're not testing the result.

What you present as a tangent is exactly testing if the method works. And that's not a tangent, but crucial to knowing its accuracy.

Okay, I guess you really want to go on these tangents. And it's going to take a while, because we'll probably need to do at least three or four of them to get out of here.

If the claim was that in every case we had evidence, the proposition we had evidence for turned out to be true, then yes, we could test by comparing to known quantities. But that's not the case. We can have evidence for conclusions that turn out to be false, we can have conflicting evidence. If evidence guaranteed truth, it would be called proof.

So we can't test for whether something counts as evidence by whether what it supports turns out to be true. Not on a case by case basis.

 

[Cavemonster]

The likelyhood might change the amount or nature of evidence that would be needed to support the existence of the lawyer.

The likelihood IS how certain we can be that this lawyer exists.

I feel like we may be butting heads on semantic rather than substantive points, but I'm afraid I don't know how to clarify them.

Could you explain to me the difference between these conclusions below.

On the question of how likely it is that a person named Daniel as described in other posts exists, the evidence leads us to believe it is nearly certain that he does.

On the question of whether a person named Daniel as described in other posts exists, the evidence leads us to believe it is nearly certain that he does.

I don't see any functional difference between the two.

 

[Mojo]

The likelihood IS how certain we can be that this lawyer exists.

No, it's how certain we can be that some lawyer called Daniel might exist. It isn't evidence for the existence of any particular lawyer called Daniel.

 

[Last of the Fraggles]

I'd say it's more than fairly likely, it's guaranteed.
I would bet money any day of the week that a person fitting this description existed.

You say these are known quantities? How are they known? The same sort of evidence that I pointed to in my first post.

We know that New York has a high population.
We know that a significant percent of that population is Jewish.
We know NY is an urban area with a lot of law firms.
We know that Daniel is a very common Jewish name.
We know that Jewish people not uncommonly go into law.

Now I know that right now you seem to object to the relation of this kind of broad statistical data to claims about a specific individual. Putting that on hold for the moment though, would you concede that the above is the sort of evidence you used to come to the conclusion that the claim?

"There is at least one Jewish man named Daniel living in NYC who practices law".

In this context at least, will you concede that this is evidence?

If not, by what evidence did you arrive at the conclusion that this claim is fairly likely?

I won't go so far as to say it's guaranteed but its highly likely. And the conclusion is reached more or less on the basis you point out.

But again, I reiterate, these things are known quantities (or at least we can take a good stab at them) We know there are Jewish people living in New York many of whom are lawyers and many of whom are called Daniel. You could probably do the math and work out the exact probability if you wanted to.

Nor does it it make any story about a Jewish lawyer named Daniel in New York more likely to be true.

If you say its 100% likely that there are Jewish lawyers named Daniel in New York it doesn't make it more likely that my son is a Jewish lawyer named Daniel in New York.

I'm not trying to be flippant or patronising but I seriously want you to walk through the logic.

I have a son - let's say that's a 50/50 statement right now.

I have a son who is Jewish - does the fact that many people are Jewish make this statement more likely to be true?

Your argument is that 'I have a son who is Jewish' becomes say 51/49 true because I know Jewish people exist. But if you follow that logic I can add 50 superfluous facts about my son and get to 100% true.

But I don't have a son. I made him up.

 

[HansMustermann]

That's another excuse that comes up every time: 'I'm not claiming binary true/false logic, so it's ok to talk bollocks.' So what makes you think that this time it would work?

As I mentioned repeatedly before, I'm ok with probabilistic reasoning too, as long as it's sound. In fact it even makes sense to go that route.

If you want to talk probability instead, what we have here is a P(X|Y), i.e.,

the probability that
X="a given character or event is true"
conditional that we know that
Y="it's mentioned in conjunction with a real city"

But even then the negatives count. When you divide the cases where X is true by the total cases, you can't include in the total below the fraction lines only the cases where X is true. That's a biased sample.

And if you look at the sheer mass of fiction books (or stories, as you wish) that use real cities as setup, vs the true historical books (or stories, if you prefer that) that use real cities, in fact the former vastly outnumbers the latter. If you also make it conditional of it being some demigod story, then it looks even bleaker. Whether it's Hercules or Theseus or even the NT, we have more events that can't possibly be true, yet are mentioned in conjunction with real cities, than things which could even be true, even if they mention the same cities. You have more clearly made up stories like Hercules slaying the hydra of some real Greek city, or Jesus raising the dead in a real Jewish city, than stories that can even be true mentioned in those cities.

So applying that criterion, far from moving the prior UP from 50-50, it's actually moving it DOWN.

If all you have is that a story is set in a real city, then you're not coming even near to it being more likely true than false. You need more stuff to get there.

The best you can say is that it's more probable than something using a fictive city, but that isn't saying much. That probabilities are greater or equal to ZERO is already a given, not something that needs further proof. It's like saying that something is more likely to be real than my imaginary cat. So what?

 

[Last of the Fraggles]

Okay, I guess you really want to go on these tangents. And it's going to take a while, because we'll probably need to do at least three or four of them to get out of here.

If the claim was that in every case we had evidence, the proposition we had evidence for turned out to be true, then yes, we could test by comparing to known quantities. But that's not the case. We can have evidence for conclusions that turn out to be false, we can have conflicting evidence. If evidence guaranteed truth, it would be called proof.

So we can't test for whether something counts as evidence by whether what it supports turns out to be true. Not on a case by case basis.

Lets try another angle of attack to see if it helps. Rather than trying to build up a case for the story being true. Let's work the opposite angle, let's see if we have evidence to reject the hypothesis that this story was just made up.

Does the existence of Jerusalem, carpenters, donkeys, crucifixes, Jews and sand make it any less likely that the story is a fiction? Surely the answer is no.

 

[Cavemonster]

I'm not trying to be flippant or patronising but I seriously want you to walk through the logic.

I have a son - let's say that's a 50/50 statement right now.

I have a son who is Jewish - does the fact that many people are Jewish make this statement more likely to be true?

Your argument is that 'I have a son who is Jewish' becomes say 51/49 true because I know Jewish people exist. But if you follow that logic I can add 50 superfluous facts about my son and get to 100% true.

But I don't have a son. I made him up.

I thought we could skip this tangent, but I guess we're doing all of them between you and Hans.

What you're doing is backwards from what I'm trying to explain.

As you add characteristics to a claim. Generally a probability goes down.

What are the odds of a random person having a child? Let's say I know nothing about this person, just someone picked randomly. Pretty high.
What about having a son? Around half what we started with.

What about brown haired son? The probability gets lower.

A brown haired son who plays cricket? Even lower

So you see, each addition to the CLAIM makes that claim statistically less probable. But I'm not talking about adding complexity to claims. I'm talking about how a static claim compares to real world evidence.

So of these two claims:
A man juggled fifteen eggs on Tuesday
A man juggled fifteen eggs on Tuesday in Toronto

The second one is statistically less likely, and I think you misinterpreted my point to be that it would be more likely. That's not the case.

My broad case is this.

If we start with this claim:
A man juggled fifteen eggs on Tuesday in Toronto.

Then learning that there is an egg juggling school in Toronto is evidence in favor of the claim.

 

[Last of the Fraggles]

I thought we could skip this tangent, but I guess we're doing all of them between you and Hans.

What you're doing is backwards from what I'm trying to explain.

As you add characteristics to a claim. Generally a probability goes down.

What are the odds of a random person having a child? Let's say I know nothing about this person, just someone picked randomly. Pretty high.
What about having a son? Around half what we started with.

What about brown haired son? The probability gets lower.

A brown haired son who plays cricket? Even lower

So you see, each addition to the CLAIM makes that claim statistically less probable. But I'm not talking about adding complexity to claims. I'm talking about how a static claim compares to real world evidence.

So of these two claims:
A man juggled fifteen eggs on Tuesday
A man juggled fifteen eggs on Tuesday in Toronto

The second one is statistically less likely, and I think you misinterpreted my point to be that it would be more likely. That's not the case.

My broad case is this.

If we start with this claim:
A man juggled fifteen eggs on Tuesday in Toronto.

Then learning that there is an egg juggling school in Toronto is evidence in favor of the claim.

You're right and I was a bit sloppy in dealing with this as I was hoping you would skip over it. The bit you highlight is an irrelevance to the argument so you can ignore the effect you mention.

Maybe its better to at each stage re-assess the likelihood of the original claim that I have a son given the new information

So, I have a son = x% probability

I have a son who lives in New York; OK New York is real so the statement 'I have a son is now x+y% probability'

 

[Last of the Fraggles]

In reality, I think it might be fairer to say that we first need to define what we mean by Historical Jesus but if we include 'is the person that he Jesus myth is based on' then i think we have to say the probability mathematically at least is 'unknown'

The reality of Jerusalem, Jews, Carpenters, and Donkeys does not change the result and it's still 'unknown'.

If you found something in the definition of HJ that was definitely false then you could say its been disproven but then all that would happen is the definition would be changed...