The Bigelow Institute for Consciousness Studies (funded by the very wealthy businessman Robert Bigelow) sponsored prizes for the best essays attempting to answer the question of what is the best evidence for life after death. My long blog post here reviews some of the essays that won prizes in this competition. One of the essays on the Bigelow Institute site is the essay "Advancing the Evidence for Survival of Consciousness" by Arnaud Delorme, Dean Radin, and Helané Wahbeh. Below is a critique of this long essay.
Before making some suggestions on how to do experiments that might provide evidence for life after death, the essay presents some poorly-conceived ideas about what is and is not good evidence. On page 9 of the essay we have a misconceived "pyramid of evidence" attempt to rate the strength of different types of evidence. It gives these evidence strength ratings, from strongest to weakest:
Strongest
Meta-analyses
Systematic Reviews
Independently Replicated Experiments
Pre-registered Experiments
Planned Experiments
Exploratory Experiments
Case studies and field investigations
Anecdotes
Weakest
As someone who has for quite a few years been a critic of flaws in experimental neuroscience and other types of science research (for example my post here), I consider the "pyramid of evidence" presented above to be misguided and fallacious. Here are some points to consider:
(1) Although experimental science has a certain glamour in the world of science, experimental results are not in general better evidence than observational results that are not experimental. Non-experimental evidence can be the strongest type of evidence. For example, humans observe certain characteristics in the weather. Such results are the best type of evidence, although they involve no experiments.
(2) For certain types of experiments, it is very desirable for the experiment to be pre-registered, in the sense of stating an exact research plan and hypothesis to be tested, before any data is gathered. However, pre-registration is merely one of quite a few features that good experimental research should have, and pre-registration by itself does nothing to guarantee a fairly reliable result. You can have pre-registered experiments that are utterly unreliable because of things such as inadequate study group sizes and poor measurement methods.
(3) A use of either a meta-analysis or a systematic review is no guarantee at all of high-quality evidence. In fields such as biomedicine and neuroscience, there are many very bad and unreliable systematic reviews and meta-analysis examples. You have a "garbage in, garbage out" situation. If a meta-analysis or systematic review is done on poorly designed Questionable Research Practice experiments, then the result will not be reliable evidence. And, sadly, Questionable Research Practice experiments are very abundant in certain fields such as neuroscience. In some fields a meta-analysis or a systematic review will tend to have a strong bias that reflects the dogmas and prejudices prevailing in such a field. John P.A. Ioannidis, professor of medicine and data scientist, has stated this: "The production of systematic reviews and meta‐analyses has reached epidemic proportions. Possibly, the large majority of produced systematic reviews and meta‐analyses are unnecessary, misleading, and/or conflicted."
(4) It is a huge mistake to belittle non-experimental observations by calling them all "anecdotal." Some of the strongest evidence consists of observations not taken in any experimental protocol. Jails are filled with countless thousands of people sentenced to long prison sentences based on sound observations that were made outside of an experimental framework.
(5) There are innumerable ways for experimental research to go wrong, and my long post here discusses 50 types of Questionable Research Practices engaged in by experimental researchers. Judging whether research is good or bad is a vastly more complicated affair than the simplistic little scheme suggested by the pyramid graph in the essay "Advancing the Evidence for Survival of Consciousness."
The essay "Advancing the Evidence for Survival of Consciousness" then proceeds to give us a Table 1 which gives criteria as misguided as the criteria in its pyramid diagram. The authors senselessly claim that only "blinded" and "pre-registered" evidence is strong evidence, which is equivalent to making the senseless claim that only experimental evidence is strong evidence. The same Table 1 makes the silly assertion that things observed by more than 100 people should be merely called "suggestive evidence" and that anything reported by fewer than 100 people should be called "unclear or conflicting evidence."
The evidence standards stated by Delorme, Radin, and Wahbeh are misguided and confused, and if we used their standards we would have to end up freeing most of the people who are now in prison, and also disbelieving or doubting very many of the solid claims that scientists assert as matter-of-fact realities. The essay "Advancing the Evidence for Survival of Consciousness" then proceeds to give school grades to various types of evidence for life after death. Most of the grades given make no sense, because they are based on erroneous claims about what qualifies as good evidence.
The essay "Advancing the Evidence for Survival of Consciousness" concludes by suggesting some ideas for experiments that might provide additional evidence for life after death or the paranormal. We have this suggestion regarding experiments with water:
"(I) Apparition in the lab
This experiment tests if a deceased person could influence a physical system to reveal their presence. The physical system would be a controlled source of steam (or smoke) recorded by a high-quality camera. Steam would be used to provide a presumably easily manipulated substance that an apparition might be able to influence. A medium would be asked to invite a deceased person to influence the steam so that it appeared in the form of their face, or someone else’s face, or any recognizable shape. The picture frames of the resulting video would be analyzed using machine vision techniques to automatically identify frames in which faces or shapes appeared. Those video frames would be compared to control session frames, in which the medium did not summon a deceased person. Positive results would include reliably produced faces or shapes in the mediumistic conditions as compared to control sessions."
This would make an interesting experiment to try. But there is no particular reason to think that it would succeed, because no such experiment has previously succeeded (to the best of my knowledge). What would make far more sense is to do additional experiments using a protocol that has already succeeded in abundantly producing results "in which faces or shapes appeared" with extremely massive repetition of the anomalous results. That protocol does not involve photographing steam, but involves photographing water drops falling against a black featureless background.
The protocol is described in very exact detail in my post "How to Photograph Paranormal Patterns," which you can read here. Anyone willing to expend about 600 dollars for a camera as good as the Sony A6000 can perform the experiments. There is no need for a medium.
Following this protocol I was abundantly successful in producing results "in which faces or shapes appeared," as anyone can see by finger-swiping through the free online book here, or by examining the same photos on this site by using the link here, and continuing to press Older Posts at the bottom right. For a subset of the photos producing face-like shapes in great abundance, use the link here and continue to press Older Posts at the bottom right.
An example of one of the photos that I took on December 5, 2019 is shown below:
Below is another photo from the same day:
The pattern below (resembling an exultant human figure holding his hands high) appeared in more than 1500 photos one day, and is merely one of very many types of strange patterns appearing with enormous repetitions on individual days:
Besides many thousands of published photos showing effects like the ones shown above, I have published hours of video footage showing my camera viewfinder as I was photographing such photos, which you can see in the posts here.
The other suggestions by Delorme, Radin, and Wahbeh for experiments are a mixed bag. Some might be worth trying, and others are impractical. In general, the best rule about trying to get evidence for the paranormal is to try and do more of what has already succeeded, and to try and get more evidence of something already reported. Your chance of getting additional examples of a phenomenon that others have already reported are much greater than your chances of getting examples of some type of phenomenon that no one else has observed. There are innumerable classes of paranormal or anomalous-seeming phenomena that have already been reported. They are discussed in my long post "120+ Types of Paranormal or Anomalous Experiences," which you can read here. Trying to get additional evidence of one of those types of phenomena is a better bet than trying to get evidence of some never-observed phenomena such as deceased spirits influencing random number generators or AI programs (something suggested by Delorme, Radin, and Wahbeh).
The trio has a suggestion on a reincarnation experiment, but a bad one that is impractical. I can give a much better suggestion for a reincarnation experiment. People could be hypnotized, and asked about past lives. Any accounts they gave could be recorded, but before being awoken from hypnosis the people would be told that upon awakening they should not remember anything they said during the hypnotic regression. When the people were awoken from hypnosis, they would not be informed of what they said. Three months later the same people could by hypnotized, and asked about their past lives. A check would then be made about how well their latest accounts matched with their accounts from three months ago. I am aware of no researcher who has tried this protocol.
The paper discussed seems like one of the lesser efforts of Dean Radin, who has produced some much better papers relating to the paranormal, some of which are listed here.
The question of how to judge the quality of evidence is a complicated one, something too complex to be handled by simple little pyramid graphs like given in the paper discussed above. A better approach might be one using a checklist, in which evidence could be given a score between 1 and 15, based on how many features of best-quality evidence the evidence had. There would be separate checklists for experimental evidence and non-experimental evidence. Some of the items on the checklist would be:
- Observer is identified with his real name
- Date of observation is specified
- Report of observation written on same day of observation
- Report of observation published very soon after observation
- Photograph of observation provided
- Multiple photographs provided
- Multiple named witnesses
- Reported phenomena observed many times
- Observation was straightforward, not the result of some complex processing pathway
- Witness has no financial or career motive to slant or misstate results
No comments:
Post a Comment