Gravitational Attraction
What would happen if two people out in space a few meters apart, abandoned by their spacecraft, decided to wait until gravity pulled them together? My initial thought was that …
A long discussion about miracles
Here I summarize the (9 hour!) conversation that I was a part of here: Bad Apologetics Ep 20 - Miracles & even MORE even NEWER Resurrection evidence where we respond to the Unbelievable? episode: Is there medical evidence for miracles? Craig Keener, Michael Shermer & Elijah Stephens and a video with Sean McDowell interviewing Craig Keener called Latest Evidence for Modern Miracles. I want this post to be the place where I put as many of the specific miracle claims, the actual evidence presented, and some of the responses. I wrote about some of these cases in another post and I've written about miracles many times.
Specific Miracle Claims Covered:
- Bruce Van Natta who had a truck fall on him, destroying much of his intestine, but shortly after prayer his intestine miraculously regrew.
- Case Report of gastroparesis healing: 16 years of a chronic syndrome resolved after proximal intercessory prayer
- Chris Gunderson who was born with an incurable disorder called Chronic Pseudo Obstruction Syndrome, which was cured by prayer
- Case report of instantaneous resolution of juvenile macular degeneration blindness after proximal intercessory prayer - probably Marolyn Ford, but it can be hard to tell without the names on the papers.
- Study of the therapeutic effects of proximal intercessory prayer (STEPP) on auditory and visual impairments in rural Mozambique
- Delia Knox healed from being paralyzed after 22 years
- Sean George comes back after a heart attack
- JP Moreland gets a pool table
-
Barbara Snyder (Barbara Cummiskey Snyder) – healed from multiple sclerosis
-
The institution that is referenced for the work is Global Medical Research Institute (GMRI) and the documentary about miracles referenced is Send Proof.
Some initial thoughts
James is right that the evidence on a lot of these cases is really really bad. You can see why these stories are powerful and you can see why people really get motivated by them nd so i think it is important to respond to them because of that. The way people like the stories highlights their way of thinking and provides a teachable moment about like why don't we trust claims just from the claims, even if it has a peer-reviewed journal article. Why don't we just simply believe.
Personally I like looking into the details of miracle claims because they spotlight the scientific process, when it is misused or when people try to use the authority of science but don't want to go through the admittedly hard work to earn it.
Case studies as the basis of evidence
The primary studies here are Case Studies, which are some of the lowest and least informative types of studies in medicine (see https://guides.library.stonybrook.edu/evidence-based-medicine/levels_of_evidence) with Meta-studies, Systematic reviews, and Randomized Controlled Studies being the most effective forms.
Case studies may be suggestive, but they are there so that you can propose hypotheses to test more rigorously. No one in medicine evaluates the effectiveness of a treatment from case studies alone, or even primarily due to case studies.
Fishing for positive outcomes
On GMRI's website there is the following request:
"We’re always looking for new testimonies. Do you have a testimony that is medically verified?”.
Could you imagine a similar claim written about, say, a homeopathic remedy for cancer? If you took our remedy, and improved, we want to hear from you. Otherwise, not. Regardless of the contents of the specific studies, this sort of fishing for positive results is thoroughly unscientific and casts significant doubt on any results from this institution.
Worldviews and Evidence
Even if miracles could be the case, regardless of your worldview, how would you be able to demonstrate that God did it? Once you admit there are other agents, you need a process for actually determining that God -- as opposed to some other agent or natural process -- actually was the cause.
I also think that there is a two-stage process for investigating these and similar processes (i.e. aliens, homeopathy). The James Randi foundation did this as well, with a preliminary test before the more in-depth test. The point is to first establish that there is some interesting effect there, regardless of the cause, and only after passing that threshold do we get to consider a more in-depth investigation.
Also, the theists show a profound lack of imagination in coming up with studies that can test the involvement of an agent, while ruling out other effects. James Fodor suggests, as an example, doing the prayer studies in different languages to help rule out some of the competing effects. The fact that someone can come up with a simple idea like that, and none of the studies include these sorts of considerations, show that the theists aren't even trying to test their ideas.
Base-rates
The entire approach of those arguing for miracles focuses on the (presumed) successes but in any real study for a treatment, we need to know all of the cases -- the ones that worked and the ones that didn't -- to provide evidence of efficacy. Also we need to know what happens when there is no treatment as a control. None of these steps are taken in any of these cases.
Physicians
Keener quotes a study that showed that half of the physicians studied said that they had seen what they consider to be miracles. However studies also show about 80% of ER nurses and 64% of ER doctors believe lunar cycles affect patients' mental health, even though the data do not support such an effect. There are any examples of this, so see https://pubmed.ncbi.nlm.nih.gov/15166467/ and https://nationalpost.com/news/canada/full-moon-does-not-affect-mental-health-emergency-room-study-suggests as a couple.
Controls
James Fodor compares healing claims to homeopathy -- we have millions claiming an effect, and if we just counted those then we might be convinced that homeopathy was a real treatment. What is missing are the controls -- how often does the "treatment" happen and there is no effect, and how often is there an effect but no treatment. A comparison of base-rates is needed and Keener doesn't seem to care about this when it comes to miracles.
Timeline and a Face Palm
James rightly points out that the most these people ever seem to provide is:
- evidence that someone was sick at one time
- evidence that they were not sick at a later time
- sometime between those two points they claimed that they were prayed for (and don't always demonstrate that they actually were prayed)
The timeline definitely matters, especially if you're claiming a causal relationship. I would add that most of these people are praying most of the time, so it's kind of like looking for astrological causes behind earthquakes -- since earthquakes happen all the time, you can always find one preceding or succeeding any other event you want.
How do we know the initial diagnosis was accurate?
Bruce Van Natta
- Bruce Van Natta who had a truck fall on him, destroying much of his intestine, but shortly after prayer his intestine miraculously regrew.
From the Send Proof film they admit that they can't publish Bruce Van Natta case because the accurate measurement of the length of intestine needed to verify the result we don't have the technology yet. The current results are suggestive, but are not conclusive. This means that the original measurements were not accurate enough to establish the lengths, and the follow-up that said it grew back are also not enough.
- they used xrays to diagnose it originally, which is not the best for soft tissue
- perhaps a thickening of the intestine would look like a regrowth?
- the natural explanation -- some natural healing plus the imaging being unreliable
- the secondary measurement said "approximately 6 feet" as opposed to 3ft, so the "approximately" is doing a lot of work
- perhaps under extreme damage there is more healing, but we haven't seen this in other cases because of the rarity of the effect
James:
The trick is you have to dig a lot to get the actual facts of the matter, they don't typically present these to you like "here's the facts, and this is why we think it is a miracle" it's "the guy's intestine regrew" but that's not an established fact at all that's their interpretation of a certain set of facts. It makes it very hard to investigate these claims because you have to spend so much time trying to work out what is actually established.
This is like the difference between theists asking skeptics to explain how the disciples ate fish with Jesus, when in reality what we have to explain is that we have a text that has a story where the disciples ate fish with Jesus. Taking the story as the evidence is problematic.
Case Report of gastroparesis healing
The story is that this person had a condition where they couldn't eat normally, were fed with a tube, and this persisted for 16 years. Then he was prayed for and was instantly healed, could eat, and have his tubes removed.
An immediate thought is how many times was he prayed over and nothing happened? This goes a long way of establishing (or not) a causal link.
There's a long list of reasons for this poorly understood condition. Also, this condition is known to come and go in some patients. These are two red flags which will need some level of caution in establishing the cause of the condition improving.
The timeline of the events from the paper is
This explores some of the rarer causes, but more importantly, the study they reference for the viral cause for the condition had cases that resolved in as long as 2 years (N=11).
- perhaps this was the case here either with a longer resolving time (N=11 is not that large)?
- or perhaps it resolved in 2 years, and they never checked for a long time, only checking right after the prayer 16 years later.
This is the reason the specifics matter. We know that the healing was observed after the prayer, but we don't know if it caused it because we can't rule out an earlier recovery that went unnoticed. Even the supplementary information doesn't rule this out, where they have some reports in the 16 years but don't say they've been attempting oral feedings. We also can't rule out that he was prayed for many many times in 16 years, and only at this particular time for some other reason he recovered. It is for these sorts of problems that randomized controlled studies are developed and are necessary to distinguish effect from non-effect.
This would certainly be testable. I think, however, there is a selection bias here -- those people who have theological reasons to expect healings are more likely to broadcast what they consider to be healing events, while others for the same sorts of events will prefer more modest interpretations. Thus we will more often hear from the healing crowd.
Craig Keener at around 37:00 in the Unbelievable? episode says
if i'm not mistaken the one that Herbert Benson conducted was that when they were trying to recruit people they were having trouble getting certain participants that all the protestant participants were from the unity school of christianity which actually doesn't believe in miracles and so you know they had a limit it's not the ideal way to conduct a prayer study
later, Keener continues,
if there's a pattern like nine times out of ten where this happens it's in the context of prayer that that should say something again we we need a whole lot of a whole lot of cases to be able to to examine that
These are both testable, yet there seems to be little effort by theists to improve the studies, they seem to rather prefer falling back on case studies. This shows a profound lack of imagination on the part of the investigators, and I think shows their hand. They want the positive results and the authority that comes from the scientific enterprise, yet they aren't willing to put in the real work to earn that authority.
In Steven Novella's post at Proximal Intercessory Prayer, he comments (on a different case, but applicable here),
But there is another reason to study “proximal” prayer, which in this study meant physical laying on of hands, not just being near the subject. Studies of distant intercessory prayer have been essentially negative – we see the typical random scatter of results expected of an ineffective treatment, with no consistent pattern of positive results, and with the best studies being negative.
...and...
In my opinion this study represents a larger trend that I have discussed before – clinical research going backwards in quality after higher quality studies yield negative results. We have seen this with acupuncture – after well-controlled trials were negative, some proponents decided that “pragmatic” (i.e. unblinded) studies were better. We see this with homeopathy, where well-controlled studies are negative and then dismissed because treatments were not individualized.
The one-on-one treatments, when they show an effect, are confounded by the more intense experience, the one-on-one attention, the group pressures, etc... You are adding a lot of psychological elements which make it harder to establish causation.
Chris Gunderson
Chris Gunderson who was born with an incurable disorder called Chronic Pseudo Obstruction Syndrome, which was cured by prayer
This one is much like the previous one, but there is no study -- just the stories. We don't know the timeline at all or the specifics, or any actual measurements. We only have the stories. It's here for completeness, even though there isn't much more we can say.
I misspoke in the episode for the next one, thinking the name was Barbara Snyder, but Barbara's case is an MS case. I made a number of these errors in the episode, and I think it is because 1) the scientific papers (rightly) don't use names and 2) when the stories are told they don't cite their sources well and 3) I'm terrible with names. Sorry!
Case report of instantaneous resolution of juvenile macular degeneration blindness after proximal intercessory prayer
- Case report of instantaneous resolution of juvenile macular degeneration blindness after proximal intercessory prayer - probably Marolyn Ford, but it can be hard to tell without the names on the papers.
The story is
an 18-year-old female lost the majority of her central vision over the course of three months in 1959 and was later diagnosed with juvenile macular degeneration (JMD). In 1972, having been blind for over 12 years, the individual reportedly regained her vision instantaneously after receiving proximal-intercessory-prayer (PIP). Subsequent medical records document repeated substantial improvement.
There are many causes of this disorder.
It was assumed that Stargardt's disease was the best fit for her symptoms, but she has no family history for this genetic condition. Further, she also has a twin sister! Since these diseases are genetic, it makes it a lot less likely that she had this condition.
So her initial eyesight is 7/200, quite poor.
The emotional language in the paper is a little unusual for a medical journal, and seems a bit sensationalist. It also points to placebo as a possible contributor to the perceived magnitude of the improvement -- the amount of improvement is likely to be perceived to be higher given the expectation. This doesn't affect the doctor reports, but the personal reports.
The first doctor report is two years after the event (the event in 1972, the doctor visit in 1074)? This seems really strange, if we are to believe the magnitude of the healing -- she claims to have gone from nearly blind to nearly perfect instantaneously, yet doesn't get this checked out immediately?
The 1974 measurement is 20/100 compared to legally blind at 20/200 (uncorrected) which is still quite poor and far from being "restored", but still a large improvement from her measured 7/200 mentioned earlier.
We make a mistake in the episode here about corrected vs uncorrected vision measurements. Honestly, this mistake is primarily due to sloppy reporting -- if you are reporting a set of numbers in a study, especially if you are establishing a timeline of events, you need to make it clear when some numbers are not comparable. You don't mix, say, corrected and uncorrected measurements in a timeline unless you provide both at every time point you have. To mix them makes it much harder to determine what's going on. We don't have a real timeline on this case, so we can't tell where the causation is, when significant improvements have occurred, etc...
- perhaps a misdiagnosis of Stargardt's disease, and a gradual improvement for other reasons.
- because her twin didn't have it, it makes the original diagnosis questionable
- the technology at the time is cited to be suboptimal in the paper
- we can't tie the prayer specifically to the timeline, because of expectation effects and placebo effects, so causation is in doubt
- she had a genuine improvement, but we have to ask which is more plausible, that there was a misdiagnosis (with the extra point about the twin sister) or the miracle played out in that way
If this is the Marolyn Ford case, which I think it is, then we have to consider that she worked in a church for that amount of time, and obviously got prayed for many times with no effect, and only got the healing after many years. I could just as well say that her brushing her teeth caused the healing, because she was healed in the evening shortly after she brushed her teeth. Of course this is ludicrous, because teeth-brushing would occur twice a day, every day, so every single event would follow a teeth-brushing. Prayer, in her house, probably occurs even more frequently.
it's hard to trust someone who says things like this, when we know that the measurement afterward was 20/100 -- far from perfect. Of course, it's possible her sight was perfect and declined in the 2 years between the "healing" and the measurement which would call into question the miraculous nature. It also highlights the need for detailed timelines to establish what the effect really is, before trying to grasp at causes.
Kamil points out that there is a mismatch between the magnitude of the claim and the attention they pay to documenting it. If you thought you were miraculously cured, you'd have it tested every two weeks not waiting two years. There may be a difference between the way they look at things now vs the way they looked at it at the time of the event -- which is perfectly explainable under naturalism. This effect has been seen a lot, but a good example is the Rendelsham Forest UFO sighting where there is a big difference between the original reports and the later reports or the Barney and Betty Hill case of alien abduction where one can track the contamination of the story to popular media depictions of aliens.
I think it also is evident that the theists aren't serious about studying these things, and are content to believe things on bad evidence. It's the same reason large careful studies aren't done -- they simply don't care enough, and are already believing, so why bother?
These two cases are some of the best they have, and they are only modestly interesting, and have plausible mundane explanations. Even if we couldn't explain it, we'd say "this is interesting, maybe we should look into these effects more" -- this is what we do for any other proposed medical treatments when we have only two case studies to go on. We wouldn't conclude that these things actually happened as reported, especially given the gaps in the timelines. The fact that there are only two or three of these, that these are the best they have, given all of this time is damning evidence against their case that miracles actually occur at even the 1% level. At over 2 billion Christians, praying many times, if this was a real effect at the 1% level we'd expect millions of documented cases yearly -- certainly a large enough signal to experimentally verify. The fact that they can only produce two or three, with serious procedural and reporting issues, is strong evidence that this effect is non-existent.
The stories change
We can see the stories changing in the telling, being simplified, critical details dropped. This highlights many of the problems I've been talking about and we get to see it in real-time.
Keener on Replicability
i think michael's right about obviously so many things are coincidence so we have to weigh a lot of factors to see um you know is there a theistic context to it that that tends to predominate in in the cases where we have these and so on you need a lot of cases to be able to to test that but uh the issue that he raises about replicability is is a real issue in the sense that you know when when we speak of of god creating everything and we speak of you know as christians speak of god's handiwork being all around us you know that's that's those are things that can be tested for replicability but we may disagree on the explanation you know levels of causation but miracles as one-offs so to speak they're not replicable and so they're not really testable by replicable means you can't um you know ensure that you're going to have one the next time around at the same time there's all sorts of things we have to deal with that aren't replicable certainly in historiography i have to deal with that so we want to know how somebody died we can't kill them again to to to ensure that and that's why i brought up how in different different disciplines we have to use the epistemic approach that's suitable for that discipline and so for miracles i think a case study approach is much better
This is pretty silly. They are claiming there is a causal effect, they are claiming that these things happen regularly enough to be observed, yet they want to hide behind the idea that it is somewhat random and not actually try to test it. We don't do this for any medical claims. If someone says a treatment works, even at the 1% level, it is tested carefully to see if it works -- especially if there are no obvious side effects. This is special pleading -- wanting to hold their claims to a lower standard than other claims.
They also make the claim that, for example, in Heidi Baker's ministry in Mozambique these healing happen "all the time" and are thus replicable. It is a notable pattern for something to become less observable to more carefully you're able to observe it. UFO sightings happened all the time in the 1970s and 1980s but suddenly drop when we put cameras in everyone's pocket? This is the pattern we see of something that is non-existent, and the healings follow this exact same pattern. If something is real, when we can more carefully observe it, it becomes easier to see not harder.
Kamil makes the interesting point that the testimonies of different religions only reflect what that particular religion has as a theological component. Near death experiences are only reported within religions that support dualism. Stories of recalled past memories from past lives only occurs within religions that support reincarnation. And why aren't there miracle claims of people spontaneously generating electricity? This is totally expected on naturalism, but not expected under theism. It's not impossible under theism, but it is an odd omission and speaks to God's motivation. James points out that God seems to choose not to use all his abilities -- he could heal amputees, even while someone is in an MRI, but chooses note to. God could choose methods that are not so easily confused with memory biases, measurement errors, and other distortions and it is suspicious that he doesn't operate that way.
On Standards of Evidence
From Send Proof Elijah StephensI didn't realize how difficult it would be [to find evidence for miracle healings]. If someone gets healed, you often can't get back in contact with that person -- "it almost feels like if a miracle happens you have about 48 hours and then their gone". You need to collect the records, have a medical professional who can actually make sense of that data, and have a medical specialist who is willing to look at the data. A doctor won't want to look at a regrown kidney because it will look like he did insurance fraud. Also there is this general trend of naturalism in which you are looked down upon if you believe that the supernatural can occur so there is this whole culture at play where the medical community is working against finding truth"
I hear the theist complaining about how hard it is to test miracles and I find the objection lame. Try comparing these claims to, say, the difficulty in confirming the Higgs boson, gravity waves, black holes, or the accelerating universe. Each of these is far harder than the rather trivial challenges to testing miracle claims. Also, while we have been generally optimistic that these miracle claims are not the result of deliberate fraud, one does have to consider that as a real possibility and find ways to rule it out. Finally, this gripe about a naturalistic bias is completely misguided. The first step -- which the theists have not achieved -- is to just demonstrate that there is a real effect to be investigated. These theists are in fact arguing for a lower level of evidence than we demand of the universities and pharmaceutical companies for their claims.
We are interested in the truth and are aware of many other factors (e.g. placebo, memory contamination, fraud, spontaneous remission, etc...) that will confound the effect so scientists have set up steps (e.g. randomized controlled studies, well considered controls, etc...) to avoid these confounders. These steps are in place to get at the truth, not to specifically put up barriers to rule out miracles. If miracles are real then these steps -- while possibly inconvenient -- would be easy to demonstrate. Pharmaceutical companies don't complain about these steps, they just do the work. The theists seem unwilling to do the work and want a pass anyway. Whining that it's hard is just lame.
As James says, "God really needs to up his game."
Why are these miracles as opposed to just rare events?
Justin: why do you feel sympathetic to the view that what they're finding is evidence for supernatural healing rather than just extraordinary somewhat unexplained events ?
Keener: what is specific about that data that suggests it isn't just a fluke you know one of those one in a million things that just happens sometimes yeah anomalies occur so it's important to look for an accumulation of anomalies so you don't just have what could be a coincidence. [...] these things actually do tend to happen in certain circles more than more than in others and that's why for me it would be hard for me to see it as a is a coincidence
This is a direct claim, that these miracles occur much more frequently in some groups (e.g. Christians) than other groups. Easily testable, and I'd be surprised if this is a real effect once you get past the selection bias and sociological effects, and have good controls including separating number of miracle reports from actual anomalous medical events. For example it's obvious you'll get more miracle reports in Christian circles than, say, atheist circles even if miracles don't exist. I doubt Keener has tried to actually get this done beyond the anecdotal level. Keener doesn't explore alternative hypotheses and how he'd distinguish between them. He seems really swayed by anecdotal evidence in everything he's written or said, from my experience.
The murderer and prior experience
Shermer: well the counterfactual what about the christian families who pray for their loved one who's dying from cancer and they just die nothing happens from the prayer that happens all the time. [...] why are they not healed if the context matters and we're defining a miracle as you know divine intervention not just a highly improbable event then how come there's no divine intervention there and if the answer is well god works in mysterious ways who knows well you know that's kind of a hand wave
Elijah: i think also calling a handwave kind of undermines something is imagine you caught a murderer and uh you saw him shoot a guy and he goes well i i haven't shot anyone before in my life um is that what we're trying to show is these are real examples i can't explain the soldier coming home that that makes perfect sense but what i can say is there was a blind lady and she got prayer and her eyes opened up and she had macular degeneration they can verify that before and afterwards and i'm not sure it's rational to call those uh uh just coincidences
The analogy doesn't work, and honestly I think this is a miracle talking-point that Elijah is misusing here. When I've heard the murderer analogy it is to argue against Hume's uniform human experience against miracles, because someone could argue for the murderer by claiming it is against his uniform experience to murder someone -- he's never murdered anyone before. In this case, Shermer's argument is that there is a context (e.g. prayer, laying of hands, religious service, etc...) where miracles are claimed to be happening but they happen in only a small part of those same contexts and there's not a good reason give for that difference. The analogy breaks down here because of a difference of information (we have some context for prayer but none in the analogy) and it seems out of place.
A question also came out that if the scientific standard seems too high for these theists, in which discipline would the standard not be too high? Not legal, not medical, not historical, etc... Doesn't seem to be anything other than perhaps theology.
Inference to the best explanation
Elijah: they can verify that before and afterwards and i'm not sure it's rational to call those uh uh just coincidences like you're assuming some type of causality there um and i i think if you even allow for the possibility of something like the christian god to exist then um you've got to take seriously that it could be his action and then make an inference to the best explanation and where i think the voice of science comes in the best is in eliminating hypotheses not necessarily proving that's his voice and so you can go back and go well there's seven different hypotheses science can help us before eliminating some and then we've got two or three and we've got to critically think and go what do i think's most likely the case and i think that's the human situation and we make that call all the time we make it in court
Nathan: it's like he understands broadly what he should be doing but then that stands in place of him actually doing it it's like as if just finally having a bit of a grasp of how we should reason about the these things....suppose we agree with you now well why is that the best explanation?
James: yes so think about what Shermer has done. He's effectively asked "look, according to your explanation we would expect miracles to occur actually in many more cases because god apparently has the the reasons and the motivation and the context for him to perform miracles there but he doesn't, so there's a plausibility issue there with with your explanation" Shermer's identified one aspect that seems to be accounted for poorly on the miracle explanation. The response we get is to say "Well what you have to do is you have to compare explanations and see which one's the best" and it's like "Right! Yeah ,that's what Ii just asked you about!"...can you address the point and not just resort to this other issue that is actually not really in contention? These responses just seem so, I don't know, juvenile to me like it's like they haven't thought about these issues.
Nathan: I think what we're seeing here are the fruits of apologetics in some ways. I sort of think that these popular apologetics books will briefly touch on some of these tools that should be used but then that's a stand-in that makes people think that they're actually using them. It's like "no trust me, this is a legit methodology and it leads to the conclusion" and that's enough to legitimize your reasoning process. Just to have the words "inference to the best explanation" that legitimizes the inference that you're making rather than actually being able to use inference to the best explanation.
Nathan's point here is interesting, especially in light of the idea that the murderer example is not being used in the right place, that Elijah has this set of standard "tools" that he pulls out but doesn't seem to understand their use or limitations and is prone to misapply them.
Kamil: so he basically said, look the way we do it is that we have, let's say seven different possible explanations, six of them are naturalistic which we can disconfirm using science and then the seventh is the explanation that God did it. And the way we proceed is that we use medical science to eliminate the six and we can't really confirm or disconfirm the last remaining one but because we eliminated all of the alternatives that must be the one right or at least it's reasonable to believe that this is the one -- it's the best explanation. I'm just wondering, is it possible to do it the other way around? Is it possible to start by eliminating the theistic hypothesis? I think no, because theism is vaguely defined. How would you eliminate the hypothesis that a miracle did in fact happen?
James: I'm not a fan of eliminating explanations because i think that that's almost impossible because you can always augment the hypothesis with further details, so the issue is going to be one of comparative plausibility and explanatory scope. So for example, if I just say oh well there was spontaneous remission you could say that's implausible given our medical knowledge but it's going to be very hard to establish that that's actually like impossible, to definitively rule it out and so the question of which is a more plausible explanation and not can you rule it out.
Brian: Well actually I think you can do a bit better than that by using essentially the case studies to say, okay, these are the sorts of situations where we would expect miracles to happen and what types of miracles we might see , sometimes we have things that go into remission so how do we rule that out, we have to come up with some kind of randomized control study in order to test miracles. And I think with a little imagination one can probably control for things like spontaneous remission and background effects and even some of the kind of unknown effects by a suitably designed experiment. But theists show no interest in doing this, they're satisfied with the first bad study that confirms what they already believe.
Brian: Essentially Shermer's complaint about prayer being done and nothing happening is part what you would have in any kind of controlled study because you would have the case looking at all prayer events and which ones had effects and which ones didn't, that's the basic thing they seem unable to even grasp that idea and it shows a certain unwillingness to actually try to find the truth and to rule out these actual alternatives. It's not enough to show an effect that's really rare, maybe that's your first motivation, like "oh wow, this is a weird case. maybe we can study this" That's how a lot of science is done, it starts with "oh wow, that's weird" and then we can go and start testing it but theists don't seem to even want to take that next step.
Skeptical of your Skepticism
Elijah: you got to be skeptical of your skepticism so if you don't allow for a God to exist then you're not open-minded ... I could show you cases of arms growing back theoretically speaking and you would not believe it you would come up with natural explanations
This phrase shows a profound lack of understanding what skepticism is. Skepticism is basically that we proportion our belief to the evidence, and we've found (through science) that to do it we need to disbelieve most claims, that we need to acknowledge our biases, and to recognize that even the scientific process is limited. This last part is the reason we set up replication, controls, and other processes to minimize the effects of the limitations. This kind of phrasing is trying to stop people from being skeptical of their claims (and only their claims). It's part of the lowering of the standards of evidence. As Nathan puts it, it's telling believers who doubt "stop those naughty thoughts." James says, to refuse to put in the controls to avoid the biases we know we have, is a form of dogmatism -- we're looking for the result we already believe and we refuse to put in the methods that would allow us to isolate the different variables.
James: can we just make it a thing to always mention that like although it's not that popular these days that deism or non-interventionist form of theism are a thing like you can totally believe in god and still not be okay with miracles? It just really annoys me this is always conflated. It's just like a lack of imagination there's just like Christian fundamentalism and there's atheist naturalism and there's no other world views, apparently.
What really annoys me is when they say "I could show you cases of arms growing back theoretically speaking and you would not believe it" when the best evidence they have barely rises to the level of being interesting, let alone something amazing like this. Don't presume what I'll believe, show me evidence of an arm growing back first, then perhaps we'll see -- but I'd need the proper controls. Could it be an illusion? False reporting? Has it been confirmed? Etc.... Yes, I'll look for these mundane explanations first before jumping to agency, because those are the most common explanations of extraordinary claims. Any legitimate effect should be able to withstand this scrutiny (and more!), but with the best they ever give us (like the cases above), the second you dig a little bit you find that all the evidence kind of vaporizes or there are plausible alternative explanations that take only a minute to come up with. There is nothing even close to an arm growing back. The same thing occurs with UFO claims, homeopathy studies, and psychic abilities.
Keener on Confirmation Bias
Keener: There's confirmation bias I think that can happen either way. I think a lot of times people will see something as a miracle because we're inclined to believe in miracles, you can also have confirmation bias to say I'm inclined not to see this as a miracle because I just don't think those things happen.
The bias is on the person making the positive claim. What is being missed here is it shouldn't matter what the skeptic says -- the evidence should be strong enough to convince the skeptic. Quoting myself here (from this post),
Here are two examples from my life where I was skeptical of an extraordinary claim, one where it turned out my perspective was correct and the other not.
- In the 1990's, the first evidence for the accelerating universe came out. I remember thinking, "No way this is true." At the time, I thought their calibration of the supernovae used as standard candles was somehow incorrect. Did the researchers approach this skepticism with "oh, you just are biased against theories that propose unknown external forces, or violations of known laws"? No. Other groups repeated it, they confirmed any calibration, and came up with a theoretical structure (using the Cosmological Constant) to describe it. Then I was convinced. Was I wrong in my skepticism? Absolutely not. The response to skeptics is to bring the evidence to bear on it. If the evidence is not enough to convince a reasonable skeptic, then we can't be particularly confident in it.
- Recently, there was some data indicating possibly faster-than-light neutrinos. I had a response to this here, where I was skeptical of the result. Again, it was a group of careful scientists who had done the measurements, and had taken into account everything they could think of. I still didn't believe it. Did the researchers approach this skepticism with "oh, you just are biased against theories that propose unknown external forces, or violations of known laws"? No. Others tried to analyze the same data and the set-up, and the scientists explored other explanations. Turned out to be a loose cable. Here again, my skepticism was well placed.
It's not biased to be skeptical. It isn't irrational to demand a higher-than-average standard for extraordinary claims, no matter what. If you make such a claim, and that higher-than-average standard is not met, then you cannot be confident in that claim. It doesn't matter whether the claim is religious or scientific, the same rules apply. If you complain that the standards are too high, then you don't belong on the playing field and you don't deserve to be taken seriously.
Keener on replication
Keener: in terms of anomalies whether they're anomalies or whether they're explainable in some way it helps if there's a pattern. They're not replicable but if there's a pattern like nine times out of ten where this happens it's in the context of prayer, that should say something. We we need a whole lot of a whole lot of cases to be able to to examine that.
Nine out of ten with prayer is the meaning of replicable, so I don't know what he really is thinking when he says miracles are not replicable. Is he thinking (a strawman of) 100% of the time?
James: Keener says simultaneously that there are hundreds of millions of cases but also that none of this can be replicated. I don't understand how it's possible for both those things to be true. If there's that many cases there should be enough regularity to detect these things statistically which is exactly the tools we've developed in science to do and to rule out other factors. It's all just ad hoc special pleading.
On the one hand they use the excuse of "it's not replicable" when asked about how to test it, and on the other hand they want to also say that miracles are common -- more common than people realize. This is trying to have your cake and eat it too. Psychics do this all the time as well.
On charlatans
Elijah: I think this: the fact that there are fake miracles does not preclude the fact that there can be real miracles. Knowledge falls on a continuum. I don't think "I don't know" should be your default position unless that's really where you fall and so what I encourage people to do is read all sides, make an inference to the best explanation and go "this is what i think is most likely the case". Secondly we're talking about these things objectively, put yourself in situations where you can have experiential knowledge: pray for sick people that you know they would not be healed unless God showed up. If there's one thing I hope this film does for people that are inside of the Christian faith is that it makes them confident enough to go pray for sick people in impossible situations because that is a means to knowledge as well
James: Shermer's essential point is how do you distinguish charlatans from real miracles and the answer is well you weigh up the evidence and you look at the best explanation. Well, yeah but could we be a bit more specific? So the first part of the answer is sort of true but trivial and also not what they've been doing but then the second part contradicts that by saying "just assume that it's true anyway and start praying for people." What is that? Like that's just assuming what you wanted to establish. How is an epistemically responsible thing to do?
It is interesting the Elijah seems to understand that we need evidence to believe a claim, that we need to weigh alternative hypotheses, but he seems to have no idea what a controlled study is or why it is necessary.
To be charitable to Elijah's point, perhaps he's meaning that if you pray for someone yourself in an impossible situation and there is improvement, you can tell the difference between the fake and real miracle personally. He doesn't seem to understand the ways that people can fool themselves. Further, Nathan points out the problem is that Elijah needs to actually tell us some some epistemic story about what it is that puts people who pray in a position to know that God is the cause of the miracle happening. This is why causal connections are so hard to establish, because not only do you have to demonstrate a real effect, you have to somehow distinguish possible causes. Even if you said it had to be supernatural, how could you rule out Loki doing it, just for fun?
I don't think they have established a real effect yet, so it's premature to speak about causes.
Therapeutic effects of proximal intercessory prayer in rural Mozambique
Elijah: I also think we in the west have a bias toward poor people as not having knowledge of when something's dead or an arm is broken or something of that nature and so I think when people put themselves in situations where they pray for the sick and a whole village knows this child was born blind from birth they're 15 years old and that person sees gets their sight back and this is a different religion and that village testifies he was blind now, he sees that is a different scenario than we're talking about here and the lens of science really can't speak to that. But if you're that kid you've got to go what's more probable: an event that just randomly happened or the God that prayed I was praying to gave me my sight back, and so that is a part of the conversation. I think people sitting back and watching kind of don't take into account that this is life-changing stuff that's happening to these people.
As James pointed out, individual testimony is irrelevant. For example, if you take the population of people who go to Rome on pilgrimage each year, you can calculate the fraction of those with diseases that will go into remission. Each of those people are in the same position as the boy Elijah speaks about and will certainly attribute a God-given miracle to their healing. One needs larger than \(N=1\) to establish these effects -- which is why case studies are so ineffective at doing so. It will be life-changing for those individuals, for sure, even under the condition where there are no miracles.
A good response to this particular study is by Steven Novella at **Proximal Intercessory Prayer:
Let’s take the following hypothetical study: The study included 24 subjects who were all treated openly with the intervention in question. There was no blinding or control group – so everyone in the study, subjects and experimenters, knew that every subject was getting the treatment. The treatment involves active physical intervention with the subject. The protocol also calls for multiple interventions if initial treatments are not effective – essentially the subjects receive repeat treatments as long as possible until they report a response.
The outcome was either a change in vision or hearing. Subjects reported impaired vision or hearing at the beginning of the study and were tested with standard vision or audiology tests before and after treatment. All subjects demonstrated improvement from the intervention.
It should further be noted that the subjects in the study were not chosen from the random population but from a self-selected group that already believe in the efficacy of the treatment. Further, the authors admit, although they do not disclose the data, that previous attempts to document response to the treatment in other populations have failed.
At this point anyone with any reasonable familiarity with how to assess the quality of medical studies should see that this is a worthless study. This barely qualifies as a pilot study. It really doesn’t matter what the treatment is or how plausible it is – you simply cannot draw any meaningful conclusion from 24 self-selected subjects with no controls and no blinding.
Now, I know you can read headlines so you know this was a study of proximal intercessory prayer, but the point is – even if you assess the details of the study divorced from knowledge of what the intervention is, it should be obvious that this study tells us nothing.
This pretty much summarizes the problems with this study. It's amazing that this kind of study convinces anyone -- it violates pretty much every rule of good medicine with its multiple selection biases, lack of blinding, and lack of controls. Since this is touted as one of the best, and one of only three papers ever seemingly cited when skeptics ask for evidence of miracles, is not impressive to say the least. That fact alone is strong evidence that miracles are not real.
Paralyzed Woman, Delia Knox, being healed on video
Now we move on to a video with Sean McDowell interviewing Craig Keener called Latest Evidence for Modern Miracles.
The story is of a person who was paralyzed from the waist down due to a car accident and 22 years later they were prayed for and healed.
The first question is, are there real miracles that get recorded on video. Keener gives a lot of waffle, and then some video of the woman that is thoroughly unconvincing. The video of the miracle itself doesn't show much, and seems to show potentially abusive behavior of people lifting her out the chair, and supporting her as she attempts to walk in a far from medically controlled environment. There is a second video interview with the woman herself that reports completely different medical history, and differs substantially from the details Keener gives. She mentions a head injury from her accident not a spinal cord injury and never mentions that she was irreversibly paralyzed below the waist. She never quotes her doctors about her condition, and provides no documentation whatsoever -- there is simply nothing to work with. I've written about this before about UFO claims, that they fall into two categories:
- claims with no available data, so one can't investigate, so one shouldn't believe it
- claims with available data, where one can investigate, and it's shown that the evidence doesn't support the claim, so one shouldn't believe it
There could potentially be a category 3 where the evidence does support the claim, but I have never seen it in any that I have personally investigated, either for UFO claims or for miracle claims. This example is a category 1 miracle claim.
Given the number of miracle claims, and the bold claims of Christians throughout these videos, and the proliferation of cameras in people's pockets, there should be thousands of convincing videos of miracles. As I've said before, the fact that there are no videos of this sort is strong evidence against the veracity of miracles.
Someone in the chat asked "do all of Keener's cases differ as much from what actually was reported to what was reported by him?" Given Keener's habits of simplifying and distorting the actual original claims, and the demonstration throughout all of this that his standards for evidence are terribly low, it calls into question is entire 1300 page book on miracles.
Sean George has a heart attack and comes back
The story is that the Dr Sean George had a heart attack, a thousand chest compressions, many electric shocks, went on for an hour and half -- Keener says 5 minutes without oxygen, flat line for an hour and a half. When his wife prayed, having been told that he was gone, the heart monitor "sprang back to life" and then the doctors were able to revive him.
At least there is documentation published at https://seangeorge.com.au/my-story/medical-details/
This is clearly a rare medical case -- but with 7 billion people you'd expect some of these often just by the law of large numbers. The statements about being brain dead are all unsubstantiated -- there was oxygen both from the chest compressions and when he was hooked up to machinery. The statements of flat-line are also unsubstantiated -- there isn't evidence they were measuring it all throughout the chess compressions, etc... It's been shown that there is a high survival rate for patients with their heart rhythm responds to shocks. And even after CPR is stopped, there is an observed phenomenon of spontaneous recovery of heart activity after as much as 15 minutes, and his recovered after 10 minutes. Is this case unusual? Sure. Is it a miracle? Not at all, given that all of the facts of his case fall within the known bounds of what is possible and observed, even if rare. A document going through these particular medical statements with citations can be found here. It's walked though in a 5-part blog series:
- https://www.patheos.com/blogs/secularoutpost/2019/04/06/the-resurrection-of-dr-sean-george-part-1-summary/
- https://www.patheos.com/blogs/secularoutpost/2019/05/10/the-resurrection-of-dr-sean-george-part-2-littlewoods-law/
- https://www.patheos.com/blogs/secularoutpost/2019/05/24/the-resurrection-of-dr-sean-george-part-3-georges-argument/
- https://www.patheos.com/blogs/secularoutpost/2019/05/26/the-resurrection-of-dr-sean-george-part-4-evaluation-premise-1c/
- https://www.patheos.com/blogs/secularoutpost/2019/06/26/the-resurrection-of-dr-sean-george-part-5-evaluation-of-premise-a4/
We keep coming back to the question about how you would test for miracles. At a minimum, one has to
- establish base-rates for the natural recoveries
- document the timeline clearly
- evaluate alternative explanations -- and their prior probabilities
- include both the positive and negative outcomes of the treatment
- have good controls and experimental design to remove confounders
Each of the cases we've looked at have failed several or all of these properties of good investigation, and should not be believed. We would not believe any medical claim that didn't have these properties, so we are not having too high of a standard -- we are just refusing to lower the standard in these cases. The fact that the best theists can come up with are so readily handled should be a caution to anyone trusting their claims.
JP Moreland gets a pool table
Moreland: we asked god to give us a really good pool table just like the kind that you find in a pool hall, that isn't a cheap one that if you hit it up it shakes. And then I said "In Jesus's name, Amen" and so we moved on. About a week and a half later at my daughter Allison's soccer there were like 20 families and 20 moms and dads there and one of the fathers that I had said hi to but i didn't know him, walks up to me and he says "Hi, I'm so-and-so" and I said "I'm JP" and he said "how are you doing?" I said "I'm doing well" and he said "Well listen this may sound odd to you, but could you use a pool table?"
You should watch Pinecreek's reaction here -- it pretty much sums it up.
This one is obviously ludicrous, but I think undermines the theists claims quite a lot. Once you admit that getting a pool table after prayer is a miracle, then you've admitted that your standards of evidence are outrageously poor. It also calls into question the motives of God -- God is more concerned to give you a pool table than to cure leukemia in children or heart defects or nearly anything else. People casually talk about God helping to find their car keys or help their sportsing team win, which are not miraculous but imply the intervention of the agent and thus fall under the same umbrella. By counting these as "hits" and ignoring the deaths of starving children because they are "misses" and don't support their narrative is an indictment of the entire mode of thought.
Barbara Snyder healed from multiple sclerosis
This miracle claim never came up in the episode. I commented on the story here but there is very little to go on. Here's my comment,
Thinking about the Snyder "miracle", I do wonder why in these cases there isn't much more extensive medical documentation? It would seem that these people are constantly dealing with skeptics, and the easiest way to address it would be to get the pre- and post- medical test data rather than a summary by one doctor, after the fact. Is it possible she was misdiagnosed -- it wasn't MS, so looking for an MS recovery may be misdirected? Is it possible that there was MS and something else, in which case looking for a full recovery from MS is misdirected? Without further data, it is impossible to rule out these things, but you'd think the data would be available with something this recent?
From my reading also, MS can also go into spontaneous remission -- usually temporary -- but we don't have any follow-ups, detailed timelines documented, etc... We have stories. There isn't anything that could tie the prayer as the causal element, given that she was most likely prayed for often an we're picking out the one case before a claimed recovery.
