CONNIE SONNE, Dowser

[Uncayimmy]

There is no evidence that Connie has paranormal abilities.
In fact, there is no evidence that paranormal abilities exist.

Fine. Now take your statements into the real world. Pull 200 people off the street. Have Carl Sagan give 100 your statements. have him give the other 100 my statements, which would be "Connie doesn't have a paranormal ability. Paranormal abilities don't exist."

Then have Mr. Sagan ask all 200 if we should test VisionFromFeeling's claimed paranormal abilities. Do you think your group and my group will respond the same? I don't think they will. I think your group will be much more likely to take the "no evidence" part of your statement as leaving a window of likelihood that is completely undeserved.

Next ask them, "If several future reputable scientific studies showed that paranormal abilities existed, would you accept that paranormal abilties exist?" I bet that both groups would respond roughly the same.

So, in the grand scheme of things, I'll choose my statements over yours.

 

[dann]

Have Carl Sagan give 100 your statements. have him give the other 100 my statements, which would be "Connie doesn't have a paranormal ability. Paranormal abilities don't exist."

Then have Mr. Sagan ask all 200 if we should test VisionFromFeeling's claimed paranormal abilities.

If you make Carl Sagan comply with your suggestion, you'll have convinced at least me of both your psychic abilities and the existence of an afterlife!

 

[Uncayimmy]

If you make Carl Sagan comply with your suggestion, you'll have convinced at least me of both your psychic abilities and the existence of an afterlife!

Carl's not dead. He just went home.

 

[steenkh]

I think the issue here is what we mean when we say we are sure of something. BillyJoe seems to work with some 100% certainty that is impossible to be obtained, which is why he would never express absolute certainty.

I think most of us make categorically claims where it is implicit that we cannot be 100% certain. If I put my keys in a drawer, I am willing to claim with absolute certainty that the keys are in the drawer, but in reality I know that a number of things could make the statement untrue, such as there being an unnoticed hole in the drawer or I could absent-mindedly have put some other keys in the drawer. Whenever I make a categorical statement it is always with a very tiny uncertainty that I know is impossible to remove.

In the same way, I can safely claim that there is nobody hiding underr my bed, even though I have not looked, and I can safely claim that Connie Sonne has no paranormal abilities: I may be wrong, but the odds against me being wrong are so tiny as to be negligible for all normal purposes.

The accuracy that BillyJoe demands just makes no sense, because then I would always have to state that "there is no evidence that there is somebody hiding under my bed", or there is o evidence that my keys are not in the drawer.

When scientists use phrases like "there is insufficient evidence that homoeopathy works", they are using a more precise BillyJoeish language that at the same time signals to normal mortals that in the opinion of the scientists, homoeopathy works, and the evidence will turn up eventually. (Scientists may also use these kinds if phrases in order to avoid doing a Simon Singh "not a jot of evidence", and end up being sued for millions in a British court room).

 

[dann]

I can safely claim that Connie Sonne has no paranormal abilities: I may be wrong

No, you may not.

When scientists use phrases like "there is insufficient evidence that homoeopathy works", they are using a more precise BillyJoeish language that at the same time signals to normal mortals that in the opinion of the scientists, homoeopathy works, and the evidence will turn up eventually. (Scientists may also use these kinds if phrases in order to avoid doing a Simon Singh "not a jot of evidence", and end up being sued for millions in a British court room).

OK. So in order to avoid any confusion among the mortals, let me state quite frankly that
1) Connie Sonne has no paranormal abilities,
2) there is sufficient evidence that homeopathy does not work any better than other placebos,
3) homoeopathy ®, however, the gay version mentioned a couple of times in this debate, still awaits proper testing procedures. Any volunteers?

 

[BillyJoe]

So you are saying that a statement made about a specific claim should be held to be a statement about all possible claims that even vaguely sound similar? This is a straw man.

I am saying that a general statement must hold up in all specifics.

Sure, informal uncontrolled observation can serve as a starting point. It simply has abysmal specificity, and I think that it's okay to make reference to that.

Okay.

You tell me. How confident are you about the following (presence or absence)...Confident enough to state "X exists" or "X doesn't exist"?

There are different degrees of confidence, which is why I don't think we should be using absolute statements about the existence or non existence of things unless their existence or non-existence can be absolutely proven.

So you're willing to state that it's dead, just not in those exact words? Isn't this just about your own personal preference, rather than any difference in meaning?

The meaning is not the same, and I think history teaches us to be cautious.

It depends upon whether you are trying to rule-in (a significant result rules it in) or rule-out (a significant result rules it out) the idea. In the first case, your null hypothesis is the opposite of what you are trying to rule-in. In the second case, your null hypothesis is the same as what you are trying to rule-out. etc

Okay, I thought you always try to disprove what you believe to be true. The more you fail, the more likely that the thing you believe to be true is, in fact, true.

The Challenge is an example of ruling-in. A positive result effectively means that Connie has paranormal abilities, so your null hypothesis is that Connie doesn't have paranormal abilities.

An example of when you want to rule-out an idea would be ruling-out the presence of bias in the ganzfeld data or non-inferiority drug trials. The null hypothesis is that bias is present in the ganzfeld data or that one drug is not inferior to another. (Note, ruling-out studies are often erroneously performed or analyzed as ruling-in studies).

Technically, null hypotheses should be formed independently of prior probability. Rather they should reflect your goal (are you trying to be confident in a positive result or in a negative result).

But wouldn't your goal be to disprove what you believe to be the case. And what you believe to be the case would, I assume, have high prior probabilty. Otherwise why would you believe it to be the case?
In both of your examples the null hypothesis reflects what you believe to be the case: "Connie does not have paranormal ability" and "bias is present in the ganzfeld data". Both have high prior probability.

Then why can't it be considered a test of how she arrived at her claims?

Firstly, because Randi is on record as saying he is not interested in how they do it, just if they can do it.
But we can have all sort of ideas about how she achieved her results in uncontrolled tests, but I don't see how the test that was preformed could have been a test of how she arrived at her claims.

BJ

 

[BillyJoe]

Fine. Now take your statements into the real world. Pull 200 people off the street. Have Carl Sagan give 100 your statements. have him give the other 100 my statements, which would be "Connie doesn't have a paranormal ability. Paranormal abilities don't exist."

Then have Mr. Sagan ask all 200 if we should test VisionFromFeeling's claimed paranormal abilities. Do you think your group and my group will respond the same? I don't think they will. I think your group will be much more likely to take the "no evidence" part of your statement as leaving a window of likelihood that is completely undeserved.

Next ask them, "If several future reputable scientific studies showed that paranormal abilities existed, would you accept that paranormal abilties exist?" I bet that both groups would respond roughly the same.

So, in the grand scheme of things, I'll choose my statements over yours.

A practical advantage does not entitle you to lie.
Apart from anything else, you are likely to be found out.

BJ

 

[BillyJoe]

I think the issue here is what we mean when we say we are sure of something. BillyJoe seems to work with some 100% certainty that is impossible to be obtained, which is why he would never express absolute certainty.

I have always made it clear that I mean this from the point of view of scientific debate.
I'm as practical as anyone else in general conversation.

I think most of us make categorically claims where it is implicit that we cannot be 100% certain. If I put my keys in a drawer, I am willing to claim with absolute certainty that the keys are in the drawer, but in reality I know that a number of things could make the statement untrue, such as there being an unnoticed hole in the drawer or I could absent-mindedly have put some other keys in the drawer. Whenever I make a categorical statement it is always with a very tiny uncertainty that I know is impossible to remove.

I am no different.

In the same way, I can safely claim that there is nobody hiding underr my bed, even though I have not looked,

Hey? Why wouldn't you look?

and I can safely claim that Connie Sonne has no paranormal abilities: I may be wrong, but the odds against me being wrong are so tiny as to be negligible for all normal purposes.

No, the result of this test does not allow you to say that. It allows you to say only that "it has not been disproven that Connie does not have paranormal abilities".

The accuracy that BillyJoe demands just makes no sense, because then I would always have to state that "there is no evidence that there is somebody hiding under my bed", or there is o evidence that my keys are not in the drawer.

As I said....but, hey, I'm repeating myself again.

When scientists use phrases like "there is insufficient evidence that homoeopathy works", they are using a more precise BillyJoeish language that at the same time signals to normal mortals that in the opinion of the scientists, homoeopathy works, and the evidence will turn up eventually.

But it's just a matter of how it's phrased, isn't it?
They could equally, and probably more accurately, say:
- homoeopathy has no plausibility
- homoeopathy is not supported by well controlled clinical trials

(This reminds me of a metanalysis of acupuncture in the prevention of migraines in the Cochrane database that concluded "True acupuncture works as well a sham acupuncture" instead of "acupuncture works no better than placebo")

BJ

 

[dann]

- homoeopathy has no plausibility
- homoeopathy is not supported by well controlled clinical trials

So it has already been tested! My bad!

 

[fls]

I am saying that a general statement must hold up in all specifics.

So you are saying that rabbits appearing out of hats is an example of quantum fluctuation?

In that case, VisionFromFeelings' claims about seeing into the bodies of others must be true, since we already know that x-rays can penetrate human tissue.

There are different degrees of confidence, which is why I don't think we should be using absolute statements about the existence or non existence of things unless their existence or non-existence can be absolutely proven.

You didn't answer my question. Would you have the confidence to say "X exists" or "X doesn't exist" for any of the ideas on the list I provided?

The meaning is not the same, and I think history teaches us to be cautious.

So you are saying that we can never make use of our degree of confidence?

Okay, I thought you always try to disprove what you believe to be true. The more you fail, the more likely that the thing you believe to be true is, in fact, true.

This is an untenable position when you take into consideration that study results are meant to be persuasive to people who may be holding contradictory beliefs. The point isn't to persuade yourself of what you already believe, but to persuade others who do not already believe.

But wouldn't your goal be to disprove what you believe to be the case. And what you believe to be the case would, I assume, have high prior probabilty. Otherwise why would you believe it to be the case?
In both of your examples the null hypothesis reflects what you believe to be the case: "Connie does not have paranormal ability" and "bias is present in the ganzfeld data". Both have high prior probability.

Then this suggests that the study results can be of no use to Connie and other believers, since the wrong hypothesis is being tested. Does that make sense to you?

Firstly, because Randi is on record as saying he is not interested in how they do it, just if they can do it.

The "how" referred to above is about the magical, pie-in-the-sky mechanisms that claimants make up. The "how" I am talking about is the way in which unfettered cognitive biases and chance can conspire to form the appearance of a pattern.

But we can have all sort of ideas about how she achieved her results in uncontrolled tests, but I don't see how the test that was preformed could have been a test of how she arrived at her claims.
BJ

It allows you to discover whether she used a reliable or an unreliable way to arrive at her claims. If she suggests unreliable methods and shows unfamiliarity when reliable methods are suggested to her during the negotiations, it indicates that she has been relying on unreliable methods up to that point. That she goes on to fail when subject to reliable methods is almost irrelevant after that.

Linda

 

[fls]

No, the result of this test does not allow you to say that. It allows you to say only that "it has not been disproven that Connie does not have paranormal abilities".

But is that really what we are trying to say? Because, after all, it has not been disproven that the aether exists, or that the four humours cause human disease, yet for some bizarre reason we proceed as though it has. One would think that our perception that GPS tracking devices work or that anti-hypertensives prevent strokes must be very tenuous indeed.

Linda

 

[dann]

In that case, VisionFromFeelings' claims about seeing into the bodies of others must be true, since we already know that x-rays can penetrate human tissue.

Yes, it's just the X-ray goggles enabling you to look through clothes that still elude us!
(Why do so many women wear lead underwear?!)

 

[nathan]

Yes, it's just the X-ray goggles enabling you to look through clothes that still elude us!
(Why do so many women wear lead underwear?!)

if you get to see how leading underwear is, woo hoo!

 

[BillyJoe]

So you are saying that rabbits appearing out of hats is an example of quantum fluctuation?

Before quantum fluctuations, observations of nature would have suggested the general rule that "things do not come out of nowhere and disappear without trace". For this general statement to be absolute, however, it would have to be true in all specifics. It was - untill quantum fluctuations spoiled the game.

In that case, VisionFromFeelings' claims about seeing into the bodies of others must be true, since we already know that x-rays can penetrate human tissue.

Maybe one day the human retina will evolve a sensitivity to x-rays.

You didn't answer my question. Would you have the confidence to say "X exists" or "X doesn't exist" for any of the ideas on the list I provided?

I said "I don't think we should be using absolute statements about the existence or non existence of things unless their existence or non-existence can be absolutely proven". Off hand, the only one on the list that satisifies that criteria is the aether (defined as the absolute reference frame).

So you are saying that we can never make use of our degree of confidence?

No I 'm not and, yes, we can. We believe in things in proportion to the amount of evidence in its favour.

This is an untenable position when you take into consideration that study results are meant to be persuasive to people who may be holding contradictory beliefs.

We're doing the testing so we get to set the null hypothesis. We set the null hypothesis to what we believe to be true based on evidence and plausibility. We then help those with contradictory beliefs to prove us wrong. We try to prove ourselves wrong. That seems fair.

The point isn't to persuade yourself of what you already believe, but to persuade others who do not already believe.

I thought the idea was to try to persuade yourself that your ideas are wrong. The more you fail the more confident you can be in your beliefs.

Then this suggests that the study results can be of no use to Connie and other believers, since the wrong hypothesis is being tested. Does that make sense to you?

Not if we try to prove ourselves wrong and give Connie al the help she needs to prove it wrong (under controlled conditions).

The "how" referred to above is about the magical, pie-in-the-sky mechanisms that claimants make up. The "how" I am talking about is the way in which unfettered cognitive biases and chance can conspire to form the appearance of a pattern.

Fair enough. But it remains the case that you cannot be certain that it was only chance and bias that produced her series of apparent successes. The single negative test result under controlled conditions makes this more likely, but still not certain.

It allows you to discover whether she used a reliable or an unreliable way to arrive at her claims. If she suggests unreliable methods and shows unfamiliarity when reliable methods are suggested to her during the negotiations, it indicates that she has been relying on unreliable methods up to that point. That she goes on to fail when subject to reliable methods is almost irrelevant after that.

I know you have been pushing that argument and I understand what you are saying and it is a good point.
I'm still not certain, though, that it allows certainty.

regards,
BillyJoe

 

[BillyJoe]

But is that really what we are trying to say? Because, after all, it has not been disproven that the aether exists, or that the four humours cause human disease, yet for some bizarre reason we proceed as though it has. One would think that our perception that GPS tracking devices work or that anti-hypertensives prevent strokes must be very tenuous indeed.

I don't mind proceding as if these things are true, if there is sufficient evidence to do so. And this obviously applies more to some things than other things. But I'm not prepared to be absolute about it, unless the evidence is also absolute and incontrovertable.

 

[fls]

Before quantum fluctuations, observations of nature would have suggested the general rule that "things do not come out of nowhere and disappear without trace". For this general statement to be absolute, however, it would have to be true in all specifics. It was - untill quantum fluctuations spoiled the game.

Our statement that "things do not come out of nowhere and disappear without a trace" would have been based on observations about the world. So, you are saying our observations could serve as examples of quantum fluctuations and that rabbits appearing and disappearing out of a hat could be an example of quantum fluctuation. Unbelievable.

Maybe one day the human retina will evolve a sensitivity to x-rays.

The human retina is already sensitive to x-rays.

I said "I don't think we should be using absolute statements about the existence or non existence of things unless their existence or non-existence can be absolutely proven". Off hand, the only one on the list that satisifies that criteria is the aether (defined as the absolute reference frame).

The way that you are using aether doesn't count, since you are using it differently than anything else we have been talking about.

So what this is really about is that you don't think absolute statements can ever be made.

No I 'm not and, yes, we can. We believe in things in proportion to the amount of evidence in its favour.

Right. It just means that you choose to use the word to refer to situations which can never exist (the presence of absolute knowledge) and others use it to indicate a relative degree of certainty. As long as everyone is consistent, I see no reason for complaint.

We're doing the testing so we get to set the null hypothesis. We set the null hypothesis to what we believe to be true based on evidence and plausibility. We then help those with contradictory beliefs to prove us wrong. We try to prove ourselves wrong. That seems fair.

I thought the idea was to try to persuade yourself that your ideas are wrong. The more you fail the more confident you can be in your beliefs.

Falsification doesn't adequately describe what it is that science does. "Ruling-in" is not the complement of "ruling-out".

Not if we try to prove ourselves wrong and give Connie al the help she needs to prove it wrong (under controlled conditions).

Except that the Challenge is almost invariably set up to make it much easier for us to prove the challenger wrong than it is to prove ourselves wrong.

Fair enough. But it remains the case that you cannot be certain that it was only chance and bias that produced her series of apparent successes. The single negative test result under controlled conditions makes this more likely, but still not certain.

I know you have been pushing that argument and I understand what you are saying and it is a good point.
I'm still not certain, though, that it allows certainty.

We can't be certain about anything. I'm okay with consistency when indicating relative degrees of certainty and you're not. That clarifies it for me. I simply disagree with your absolutist stance.

Linda

 

[dann]

Before quantum fluctuations, observations of nature would have suggested the general rule that "things do not come out of nowhere and disappear without trace".

So nothing's changed, then. THINGS still don't come out of nowhere and disappear without a trace. Speaking of THINGS in this context is a very poor analogy to quantum fluctuations. Please give it up, BillyJoe.

 

[BillyJoe]

Oh my!

So nothing's changed, then. THINGS still don't come out of nowhere and disappear without a trace. Speaking of THINGS in this context is a very poor analogy to quantum fluctuations. Please give it up, BillyJoe.

So what exactly do you think quantum fluctuations are then?
Inhabitants of the spirit world or something?
Good heavens!

BJ

 

[dann]

So what exactly do you think quantum fluctuations are then?

Something along the lines of this, I guess.

Inhabitants of the spirit world or something?
Good heavens!

Feel free to invoke the gods of your choice.
The rules of quantum mechanics are very different from the rules of the universe 'as we know it' where things don't come out of nowhere and disappear without a trace. Thus whatever 'predictions' people may have made about the Newtonian universe cannot be expected to hold true on the quantum level.
You are the one who insists on referring to the phenomenon of quantum fluctuations as things - an abstraction you want to uphold in order to maintain your false analogy.
The inhabitants of the spirit world are also of your own invention. They don't appear in anything I've said.

 

[BillyJoe]

Feel free to invoke the gods of your choice.

I was suggesting that you were effectively doing this by claiming that quantum fluctuations are not "things".

The rules of quantum mechanics are very different from the rules of the universe 'as we know it' where things don't come out of nowhere and disappear without a trace. Thus whatever 'predictions' people may have made about the Newtonian universe cannot be expected to hold true on the quantum level.

That has been my point all along.
We never know what discoveries await us, so let's not be so dogmatic about what we think we know.

You are the one who insists on referring to the phenomenon of quantum fluctuations as things - an abstraction you want to uphold in order to maintain your false analogy.

And I think you have yet to support your contention that they are not.

The inhabitants of the spirit world are also of your own invention. They don't appear in anything I've said.

You can't have it both ways

BJ