The point about % of faith healings at Lourdes... Wow. I can't believe I didn't think of that and thanks for pointing it out. The math with this stuff always is weirder than it looks on the surface isn't it?
> I don’t like the supernatural/natural dichotomy because I have no idea how a supernatural cause (or effect) differs from a natural one.
The more I've thought about it, the more I think you would not be able to pull apart the supernatural/natural dichotomy at all. More on this below
> if I were to tell you that a high school football team scrimmaged the University of Alabama and won
This, however, would actually be a miracle lol.
> Let’s assume that the 100,000 number is accurate going back a century. That means that 10,000,000 people visited Lourdes looking for healing, and only 7,000 were healed. That’s 7 in 10,000 people were healed, or .07% healing rate.
So, imagine the healing rate at Lourdes had been 96% or even 100%; as long as people weren't totally prejudiced, it would presumably get studied. If we studied it over the century, we would certainly find some workable theory for how people were being healed at Lourdes because that's just how the world works. In fact, I think if this weren't the case, then miracles would be creation ex nihilo in some sense that the cause would be unidentifiable at all. This just strikes me as absurd, though, so the better thing to do is to assume that there would be an explicable description of what was going on at Lourdes. However, if that were the case, would we still call it a miracle, or would we just have a working hypothesis and then call it a natural event? I don't know what you think of that, or where the conversation goes from there.
Great comment! In this way, skeptics and humeans like myself are okay with saying that miracles and purportedly supernatural events could happen and be understood as “natural events.” But that doesn’t even necessarily mean they have to be scientific-ish events.
Take Lourdes for example. Let’s say there are conclusively healing powers, but only if you’re a faithful Presbyterian (just because I want to be a little goofy here). That would be great evidence for Presbyterianism, and I don’t think you could reduce the causal mechanisms to exclusively materialist explanations. It would also by polycausal.
The problem is that this is not how we experience the purportedly supernatural and miraculous! Again, not an a priori exclusion or circular logic, but a conclusion from the facts we have available to us.
So, is your point that if we did have evidence in that manner, we could conclude that there is both a natural cause and a supernatural cause (the Presbyterian God) as well? Also, to clarify, if at Lourdes almost everyone was being healed, but not a specific (religious) demographic, would that undermine the motivation for a supernatural cause? Or, in other words, does the religious specificity matter?
As a Humean, I would say we don’t know what the cause is, because we don’t know what the cause to anything is, so much as we formulate a probability we can be confident in. With supernatural causes, we especially don’t know how they operate. We can’t differentiate a supernatural and natural cause!
If you’ve ever read any books from Brandon Sanderson, it’s like magic systems. In his books they’re basically science, in that their effects are pretty well understood, except that their fundamental origin is magical.
With the Lourdes example, you could definitely speculate reasons and test them scientifically why people are or not healed, given a higher level of healing and other patterns. But at that point, it’s science!
well put. truth is though, as a non believer in any divinity my self, miracle waters and the rest, were never an explanation, but an answer. It s a simple answer to question which is faith based, a question that any aristotelian and possibly any one else would consider illegitimate. I even think that the apologists themselves do not believe that it is an explanation or proof. Not all of them are stupid. But the fact remains that any follower(/believer) needs some proof some of the time. Hence... 'a miracle is born'.
If dualism were true, we would expect souls to causally interact with brains. But then we could observe those interactions, study them, and formulate rules to predict when they occur and what their effects are. Therefore, we would then be able to naturally explain souls. So even if they existed, they would effectively be natural phenomena.
Oh yes. I'm being charitable with these numbers. I think I saw a YT video by Kevin nontradicath that said the median time of healing was in 1929 before any healing! The numbers I have seen are closer to 200k per year. No idea if it would be at an all time high or low.
> When we say “Ryan miraculously healed Judy” there is no additional causal explanation, no way of expanding on that sentence
It seems a miracle advocate could expand on this; such as "Ryan miraculously healed Judy through prayer" or "Ryan put his hands together, closed his eyes, spoke out loud to God that Judy's fever go away, inserted a thermometer that then measured 98F".
This explanation would still have all the other problems you described, especially lacking any predictive power. Still, "Ryan miraculously healed Judy" does embed additional explanatory content depending on the beliefs, traditions or theology of the narrator and not usually accepted on its own as a full & final theory of how Judy was healed.
But how the putting hands/prayer/etc had causal power lacks explanatory power. That, as opposed to normal accounts of causality. When we say "Joe sent a message to Ocean on substack" we can break that down to my thought process, my typing, it sending the message over cyber space, it showing up on your feed, and you reading it, but there's a difference between that and miracle sentences.
Yes definitely! I moreso mean that the miracle explanation is bad for all the other reasons you mentioned it's bad; and the ability to expand on it or not doesn't seem to me to do any work.
On the flip side a drug study could show significant results without an understood mechanism of action. "Patients took drug A and their fevers went down 100% of the time" might be where causal understanding ends, at least temporarily, while still being an accurate explanation.
So all I mean is that the ability to expand or drill down into causal explanations doesn't seem to matter one way or the other. If it did, we'd probably always end up at the currently irreconcilable junction of classical and quantum physics anyways; but that failure or gap in causal understanding doesn't propagate all the way down the chain to invalidate that Ryan punched Judy.
Well see, that's the problem though. Because when we're doing correlational scientific studies, we can typically theorize about the causal mechanism about why x->y 80% of the time (or so). Put another way, it's from the data of such studies that we can formulate theories of causality and an idea of why we saw this data.
When we say something like x led to y, we're usually saying that because we have lots of data to be confident of that causal relationship. But with miracles, we don't because we don't have that sort of understanding. Admittedly, all such correlations dissolve into mere probabilities at the quantum level, but for most of the time when we say x->y we are talking about/inferring causality, not mere correlations. Miracle talk thus overstates the data as it overstates our evidence of causal understanding and doesn't function in the same way.
The focus on predicting future data is more so (though not exclusively) for experimental science than historical science. Historical science characteristically attempts to construct narrative explanations for past event tokens, as opposed to come up with time-translation-symmetric predictions for event types.
Miracles, since they are by construction not reproducible or repeatable and not subject to the same regularities, are more like historical science than experimental science.
The data we work with for historical sciences, to the extent we can call those things sciences is much more agreed upon, and fits within causal networks we understand.
So you agree that the use of hypothesis testing in this space is misleading or disanalogous becomes the data is uncertain and the conclusions cannot be tested?
"Hypothesis testing" is a very generic term; is that the same as theory comparison, or do you have something specific in mind? I think miracle investigation is an attempt to do philosophy just the same as other instances of doing philosophy. You bring together all the relevant empirical and non-empirical evidence and compare different theories (hypotheses?) on their theoretical virtues, such as coherence, empirical support, explanatory power, explanatory scope, unification. It's similar but non-identical to doing science.
You can do this with Bayesian-type reasoning, or inference to the best explanation, or deductive or inductive arguments. It sounds like you might have an objection to inference to the best explanation as a general method (the best of a bad lot problem), but it's not really clear to me. I'm fine with any of these methods in this context. It's not really a methods issue to me but yeah maybe supernatural explanations never win for other reasons.
I do think how much of a mechanism one must supply to count as a good explanation is an interesting question that you raise. It seems like there sometimes are non-mechanistic explanations where an exact mechanism is not needed. This may be especially true if the interaction is primitive, such as something involving an object that doesn't have parts and thus is simple (which God presumably is, according to most views).
The point about % of faith healings at Lourdes... Wow. I can't believe I didn't think of that and thanks for pointing it out. The math with this stuff always is weirder than it looks on the surface isn't it?
> I don’t like the supernatural/natural dichotomy because I have no idea how a supernatural cause (or effect) differs from a natural one.
The more I've thought about it, the more I think you would not be able to pull apart the supernatural/natural dichotomy at all. More on this below
> if I were to tell you that a high school football team scrimmaged the University of Alabama and won
This, however, would actually be a miracle lol.
> Let’s assume that the 100,000 number is accurate going back a century. That means that 10,000,000 people visited Lourdes looking for healing, and only 7,000 were healed. That’s 7 in 10,000 people were healed, or .07% healing rate.
So, imagine the healing rate at Lourdes had been 96% or even 100%; as long as people weren't totally prejudiced, it would presumably get studied. If we studied it over the century, we would certainly find some workable theory for how people were being healed at Lourdes because that's just how the world works. In fact, I think if this weren't the case, then miracles would be creation ex nihilo in some sense that the cause would be unidentifiable at all. This just strikes me as absurd, though, so the better thing to do is to assume that there would be an explicable description of what was going on at Lourdes. However, if that were the case, would we still call it a miracle, or would we just have a working hypothesis and then call it a natural event? I don't know what you think of that, or where the conversation goes from there.
Great comment! In this way, skeptics and humeans like myself are okay with saying that miracles and purportedly supernatural events could happen and be understood as “natural events.” But that doesn’t even necessarily mean they have to be scientific-ish events.
Take Lourdes for example. Let’s say there are conclusively healing powers, but only if you’re a faithful Presbyterian (just because I want to be a little goofy here). That would be great evidence for Presbyterianism, and I don’t think you could reduce the causal mechanisms to exclusively materialist explanations. It would also by polycausal.
The problem is that this is not how we experience the purportedly supernatural and miraculous! Again, not an a priori exclusion or circular logic, but a conclusion from the facts we have available to us.
Thanks :)
So, is your point that if we did have evidence in that manner, we could conclude that there is both a natural cause and a supernatural cause (the Presbyterian God) as well? Also, to clarify, if at Lourdes almost everyone was being healed, but not a specific (religious) demographic, would that undermine the motivation for a supernatural cause? Or, in other words, does the religious specificity matter?
As a Humean, I would say we don’t know what the cause is, because we don’t know what the cause to anything is, so much as we formulate a probability we can be confident in. With supernatural causes, we especially don’t know how they operate. We can’t differentiate a supernatural and natural cause!
If you’ve ever read any books from Brandon Sanderson, it’s like magic systems. In his books they’re basically science, in that their effects are pretty well understood, except that their fundamental origin is magical.
With the Lourdes example, you could definitely speculate reasons and test them scientifically why people are or not healed, given a higher level of healing and other patterns. But at that point, it’s science!
I see. Okay, that makes sense.
well put. truth is though, as a non believer in any divinity my self, miracle waters and the rest, were never an explanation, but an answer. It s a simple answer to question which is faith based, a question that any aristotelian and possibly any one else would consider illegitimate. I even think that the apologists themselves do not believe that it is an explanation or proof. Not all of them are stupid. But the fact remains that any follower(/believer) needs some proof some of the time. Hence... 'a miracle is born'.
Great post!!
I have a similar gripe with dualism.
If dualism were true, we would expect souls to causally interact with brains. But then we could observe those interactions, study them, and formulate rules to predict when they occur and what their effects are. Therefore, we would then be able to naturally explain souls. So even if they existed, they would effectively be natural phenomena.
The % of healings at Lourdes is actually significantly worse given > 90% of those 7500 are pre-WWII amid lower standards of verification.
Oh yes. I'm being charitable with these numbers. I think I saw a YT video by Kevin nontradicath that said the median time of healing was in 1929 before any healing! The numbers I have seen are closer to 200k per year. No idea if it would be at an all time high or low.
> When we say “Ryan miraculously healed Judy” there is no additional causal explanation, no way of expanding on that sentence
It seems a miracle advocate could expand on this; such as "Ryan miraculously healed Judy through prayer" or "Ryan put his hands together, closed his eyes, spoke out loud to God that Judy's fever go away, inserted a thermometer that then measured 98F".
This explanation would still have all the other problems you described, especially lacking any predictive power. Still, "Ryan miraculously healed Judy" does embed additional explanatory content depending on the beliefs, traditions or theology of the narrator and not usually accepted on its own as a full & final theory of how Judy was healed.
But how the putting hands/prayer/etc had causal power lacks explanatory power. That, as opposed to normal accounts of causality. When we say "Joe sent a message to Ocean on substack" we can break that down to my thought process, my typing, it sending the message over cyber space, it showing up on your feed, and you reading it, but there's a difference between that and miracle sentences.
Yes definitely! I moreso mean that the miracle explanation is bad for all the other reasons you mentioned it's bad; and the ability to expand on it or not doesn't seem to me to do any work.
On the flip side a drug study could show significant results without an understood mechanism of action. "Patients took drug A and their fevers went down 100% of the time" might be where causal understanding ends, at least temporarily, while still being an accurate explanation.
So all I mean is that the ability to expand or drill down into causal explanations doesn't seem to matter one way or the other. If it did, we'd probably always end up at the currently irreconcilable junction of classical and quantum physics anyways; but that failure or gap in causal understanding doesn't propagate all the way down the chain to invalidate that Ryan punched Judy.
Well see, that's the problem though. Because when we're doing correlational scientific studies, we can typically theorize about the causal mechanism about why x->y 80% of the time (or so). Put another way, it's from the data of such studies that we can formulate theories of causality and an idea of why we saw this data.
When we say something like x led to y, we're usually saying that because we have lots of data to be confident of that causal relationship. But with miracles, we don't because we don't have that sort of understanding. Admittedly, all such correlations dissolve into mere probabilities at the quantum level, but for most of the time when we say x->y we are talking about/inferring causality, not mere correlations. Miracle talk thus overstates the data as it overstates our evidence of causal understanding and doesn't function in the same way.
The focus on predicting future data is more so (though not exclusively) for experimental science than historical science. Historical science characteristically attempts to construct narrative explanations for past event tokens, as opposed to come up with time-translation-symmetric predictions for event types.
Miracles, since they are by construction not reproducible or repeatable and not subject to the same regularities, are more like historical science than experimental science.
The data we work with for historical sciences, to the extent we can call those things sciences is much more agreed upon, and fits within causal networks we understand.
Yes, miracle investigation is not a historical science, but it is even more so not an experimental science!
So you agree that the use of hypothesis testing in this space is misleading or disanalogous becomes the data is uncertain and the conclusions cannot be tested?
"Hypothesis testing" is a very generic term; is that the same as theory comparison, or do you have something specific in mind? I think miracle investigation is an attempt to do philosophy just the same as other instances of doing philosophy. You bring together all the relevant empirical and non-empirical evidence and compare different theories (hypotheses?) on their theoretical virtues, such as coherence, empirical support, explanatory power, explanatory scope, unification. It's similar but non-identical to doing science.
You can do this with Bayesian-type reasoning, or inference to the best explanation, or deductive or inductive arguments. It sounds like you might have an objection to inference to the best explanation as a general method (the best of a bad lot problem), but it's not really clear to me. I'm fine with any of these methods in this context. It's not really a methods issue to me but yeah maybe supernatural explanations never win for other reasons.
I do think how much of a mechanism one must supply to count as a good explanation is an interesting question that you raise. It seems like there sometimes are non-mechanistic explanations where an exact mechanism is not needed. This may be especially true if the interaction is primitive, such as something involving an object that doesn't have parts and thus is simple (which God presumably is, according to most views).