Interview with Dr. Henry Bauer - Part 1

[Laird]

Linda, responding to both of your posts at once:

If you hate it, why go to such lengths to find something to disagree with?

No offence, but I didn't - it all jumped out at me. Why I did it is because I think that - especially on a public forum - mistakes ought to be corrected when they are discovered, whether those mistakes are yours, mine or anybody else's.

That's an odd way to put it. Why use name-calling to dismiss it?

Name-calling? Not at all: that which I pointed out is a simple matter of fact and very justifiable.

I could just as easily say that it is a quirk of the definition of logical implication that a true consequent always produces a true result, regardless of the value of the antecedent

You could, but you'd be totally missing the point.

so why would anyone would accept "p implies q" based on this entailment?

Which entailment? You've lost me here.

I think it bears emphasis that the potential outcomes of the different tests are conclusive falsifications or ambiguous verifications. And that it is "false negatives" (h-tests, t+tests) which provide conclusive falsification that your hypothesis is necessary. "False positives" (h+tests, t-tests) merely show that your hypothesis is not sufficient.

I don't take issue with any of that - and I agree that it is all backed up by the paper - except for the word "merely". I don't see why information about necessity is intrinsically more valuable than information about sufficiency. Perhaps you can explain why you do.

If you want to show that psi is necessary (which is probably what is needed to get other scientists to take psi seriously), an emphasis on h- or t+tests would be useful.

That's a rather unfortunate phrasing: "show[ing] that psi is necessary". In any case, I think I understand what you mean. I'm not sure though that it is universally the case that psi is claimed to be necessary for various effects to occur. Potentially, there might also be other causes of these effects. And I think that that's fine: it's enough for parapsychology to demonstrate that psi is one of the ways that certain effects can occur i.e. that it is sufficient, even if not necessary, for them.

But perhaps you could anyway suggest some possible -Htests and +Ttests for parapsychological experiments. I can't think of any realistic ones, in particular for +Ttests, but then again I haven't tried very hard. I can think of many scenarios in parapsychology where they aren't realistic options.

[Laird]

Do you agree, now that it's been explained to you, that looking at "p" and looking at "not q" are exactly similar in their potential effects and consequences? That it's nonsensical to say that one is an attempt to confirm the "idea" and the other is an attempt to falsify it? That the whole point of the paper you referred to was that it was necessary to look at both "p" and "not q" in order to test the hypothesis? In short, that you missed the whole point of that paper when you tried to present looking at "p" and looking at "not q" as alternative tests? And that much of the ensuing confusion on this thread is of your own making?

Just a small note: as I acknowledged in an earlier post, there is a difference between these two tests, in that only the test where we look first at the antecedent "p" can potentially confirm the rule - it is not possible that the test where we look first at the consequent "not q" can confirm the rule (because "not p implies not q" does not entail "p implies q").

So, it's not totally misleading to label the first test as "confirmatory" in comparison to the second, which is only capable of falsification... but nor is it totally justifiable either.

[Edit: As acknowledged in a previous post, I was wrong about this. There is in fact no distinction between the tests.]

Linda, responding to both of your posts at once:


No offence, but I didn't - it all jumped out at me. Why I did it is because I think that - especially on a public forum - mistakes ought to be corrected when they are discovered, whether those mistakes are yours, mine or anybody else's.

i don't think they ought to be corrected, unless they become relevant. I generally ignore mistakes so as not to weigh down a discussion. And they often work themselves out, anyways. This was a good case in point - I didn't actually make any errors, so you could have just left it alone. Now, we're off on yet another tangent because you introduced a number of irrelevancies/inaccuracies in order to 'correct' me.

Name-calling? Not at all: that which I pointed out is a simple matter of fact and very justifiable.

You called an ordinary aspect of logic "a quirk of the definition of logical implication" and implied that it could be disregarded on that basis. It is not a matter of fact that logical implications can be dismissed whenever 'everyday' logic feels like it.

You could, but you'd be totally missing the point.

Your point seemed to be that "not p" could not be used to confirm "p implies q" because it only appeared to do so because of a quirk of the definition of logical implication. That same quirk also would not allow you to confirm "p implies q" on the basis of "q". 

Which entailment? You've lost me here.

The entailment that finding "q" allows you to confirm "p implies q", given that q is true regardless of the value of the antecedent. It doesn't seem likely that 'everyday' logic would find that compelling.

I don't take issue with any of that - and I agree that it is all backed up by the paper - except for the word "merely". I don't see why information about necessity is intrinsically more valuable than information about sufficiency. Perhaps you can explain why you do.

I did so in my very next sentence, which you then quoted below.

That's a rather unfortunate phrasing: "show[ing] that psi is necessary".

It's a reference to "do not multiply agents without necessity".

But perhaps you could anyway suggest some possible -Htests and +Ttests for parapsychological experiments. I can't think of any realistic ones, in particular for +Ttests, but then again I haven't tried very hard. I can think of many scenarios in parapsychology where they aren't realistic options.

I already suggested several in this thread.

Linda

[Laird]

I didn't actually make any errors

It's OK, Linda - we all commit errors from time to time. I know I do. I'd respect you more if you were to acknowledge your errors rather than to deny that you've made them.

You called an ordinary aspect of logic "a quirk of the definition of logical implication" and implied that it could be disregarded on that basis. It is not a matter of fact that logical implications can be dismissed whenever 'everyday' logic feels like it.

OK, so, rather than ask, "What do you mean? A 'quirk' of the definition of logical implication? And the material conditional doesn't always accurately reflect what we mean by implication in everyday logic? Please explain", you'd prefer to assume that I'm wrong and that you already know that I am, even though you haven't heard my explanation? Isn't your principle to ask people what they mean when you aren't sure?

Your point seemed to be that "not p" could not be used to confirm "p implies q" because it only appeared to do so because of a quirk of the definition of logical implication. That same quirk also would not allow you to confirm "p implies q" on the basis of "q".

You're still missing the point. You're missing the point in a very intelligent way, so kudos for that, but you're missing the point nevertheless.

I did so in my very next sentence, which you then quoted below.

So you think that (non-parapsychological) scientists will only take psi seriously if it can be demonstrated that psi is the only way in which certain effects can occur? Why should this be the case? Aren't there various medical conditions which can have different causes? Do scientists then reject medicine because it hasn't proved that each effect (disease/condition/etc) has only a single, necessary cause?

It's a reference to "do not multiply agents without necessity".

If it is, then you are badly confused, because the type of necessity referred to in the paper, which you seem to be relying on, is very different from that. Why would you switch definitions in the middle of a conversation? Or is your point just to keep me engaged in what is increasingly seeming to be a pointless exchange?

I already suggested several in this thread.

I can't recall a single suggestion of yours in this thread of a possible -Htest or +Ttest for a parapsychological experiment that seemed realistic. If you would like to reprise any that you think are realistic, then please do so.

Just a small note: as I acknowledged in an earlier post, there is a difference between these two tests, in that only the test where we look first at the antecedent "p" can potentially confirm the rule - it is not possible that the test where we look first at the consequent "not q" can confirm the rule (because "not p implies not q" does not entail "p implies q").

So, it's not totally misleading to label the first test as "confirmatory" in comparison to the second, which is only capable of falsification... but nor is it totally justifiable either.

I'm afraid I don't see any difference. The hypothesis "p implies q" is equivalent to "not q implies not p". So looking at a "p" card has the same effect as looking at a "not q" card.

[Laird]

I'm afraid I don't see any difference. The hypothesis "p implies q" is equivalent to "not q implies not p". So looking at a "p" card has the same effect as looking at a "not q" card.

But remember that in all of this, q (or its negation) is only ever the consequent in these tests, even if we look up that consequent first. So we can never find, as you suggest that we could, that "not q implies not p", we could only find the reverse: that "not p implies not q". And that does not entail, and nor is it equivalent to, "p implies q".

[Edit: Please ignore this post. It is too badly confused to even bother trying to correct. And it totally misses Chris's point.]

[Laird]

Actually, just thinking about this a little more, even the first test, starting with "p", can't strictly confirm "p implies q" even if you turn over a card with "q" on it, because, strictly, in that case all you can conclude is "p and q", which doesn't entail "p implies q". Perhaps this is why the second paper to which Linda referred calls such a result an "ambiguous verification". It doesn't prove that "p implies q" (hence the ambiguity), but it's certainly consistent with it (hence the verification).

This, similarly, dissolves the question (my take on which you and Linda might rightly dispute) as to whether, in turning over cards, we must stipulate that "q" (or its negation) be the consequent rather than allowing that it might equally be the antecedent: instead, all we can say is that "q" (or its negation) and "p" (or its negation) occur at the same time, i.e., that they are connected via a conjunction operator.

So, strictly, the case to which you (Chris) refer would be "not p and not q", rather than (to which I objected) "not q implies not p" or "not p implies not q" (which, regrettably, I affirmed). Due to the same quirk in the definition of material implication to which I referred earlier, this does entail "p implies q" - but, again, I think we should be very clear that we would not come to this conclusion in everyday logic, nor in scientific logic.

[Edit: Regrettably, in the above paragraph I continue to miss Chris's point. Ouch. How embarrassing.]

In other words, any scientist who said, "I have demonstrated that at all times when no effects exist ("not q"), no psi was presumed to be present either ("not p"), and I have concluded based solely on this ("not q and not p") that therefore, the presumed presence of psi entails effects ("p implies q")" would be laughed at, and rightly so. The scientist's indignant "But... but... formal logic!" would not halt the peals of laughter.

Laird

Perhaps we're looking at this in different ways, but I'm just thinking about it as a problem in logic concerning the cards described in the paper on the p/q problem.

So for me the hypothesis "p implies q" just means "in all cases where p is true, q is also true". Viewed in that sense, there aren't any antecedents and consequents. We just have to consider which of the four possible cases ("p and q", "p and not q", "not p and q" and "not p and not q") are present in the set of cards. The hypothesis "p implies q" is true if (and only if) there are no cards with "p and not q". The same is true of the hypothesis "not q implies not p", so those two hypotheses are equivalent.

I agree that looking at a single card has the capability of falsifying the hypothesis - that will happen if the card has "p and not q" - but that looking at a single card can't confirm the hypothesis, but can only provide a piece of evidence consistent with it. Confirming the hypothesis would require you to look at all the cards with "p" and all the cards with "not q", without finding a "p and not q".

[Laird]

I agree that looking at a single card has the capability of falsifying the hypothesis - that will happen if the card has "p and not q" - but that looking at a single card can't confirm the hypothesis, but can only provide a piece of evidence consistent with it. Confirming the hypothesis would require you to look at all the cards with "p" and all the cards with "not q", without finding a "p and not q".

Perfect. Agreed. So, after all, the two tests (the first starting from "p" and the second starting from "not q") are indistinguishable in terms of what they tell us about the truth or falsity of the rule ("p implies q"), and thus - despite that I initially claimed that there was - there is no (even fuzzy) basis on which to legitimately call one "confirmatory" and the other "disconfirmatory".

[Laird]

A minor addition:

So for me the hypothesis "p implies q" just means "in all cases where p is true, q is also true". Viewed in that sense, there aren't any antecedents and consequents. We just have to consider which of the four possible cases ("p and q", "p and not q", "not p and q" and "not p and not q") are present in the set of cards. The hypothesis "p implies q" is true if (and only if) there are no cards with "p and not q". The same is true of the hypothesis "not q implies not p", so those two hypotheses are equivalent.

Agreed, in formal logic, these two hypotheses - "p implies q" and "not q implies not p" - are equivalent. i.e. (¬q→¬p)↔(p→q) is a logical (necessary) truth, where ¬ is the negation operator, → is material implication and ↔ is the "equivalence" aka biimplication operator.

However, as I wrote, and for which I provided an example above, in everyday or science-based logic, we certainly wouldn't consider them equivalent. But I'll stop banging on this drum now since nobody seems to be interested in it.