WISEMAN AND O'KEEFFE (2001) provide some critical comments about the Schwartz, Russek, Nelson & Barentsen (2001) paper that reported findings from two studies involving wellknown mediums. Their opinions involve a combination of:
1. statements made by Wiseman and O'Keeffe that do not acknowledge the fact that the same issues were previously discussed in Schwartz et al. as possible questions for future research, and
2. statements made by Wiseman and O'Keeffe about a subset of the data that selectively ignore other (and important) data reported in the paper which are inconsistent with - and seriously question - Wiseman and O'Keeffe's speculations.
I have written this response not only to correct some of their errors of selective omission, but to raise a larger question about the need for balanced "evidence based skepticism" (a term used by John Astin, personal communication). Is it possible for skeptics to provide criticism in a manner that is balanced and true to how the data were actually reported in the published research papers?
First, concerning questions raised by Wiseman and O'Keeffe about (1) possible sensory leakage, (2) the potential for judging bias, and (3) the use of an "inappropriate control group", the reader should be aware of the following discussions in the Schwartz et al. (2001) paper:
Possible Sensory Leakage: What Does it Mean?
purpose of the original Schwartz et al. experiments (2001) was not to eliminate all possible sources of sensory information, but rather to:
(a) begin with a semi -naturalistic design eliminating visual feedback and much auditory feedback (only yes / no responses were allowed by the sitters - Study 1 - this was done to develop a professional relationship with the mediums as well as to witness how mediumship is typically practised), and then to
(b) eliminate visual and semantic feedback, so that only subtle breathing, body movement, sighs, and possibly odours would remain (i.e. the sitter-silent part of Study 2 - a first step toward addressing conventional sensory hypotheses).
This was clearly stated in the paper.
The question is, could the remaining subtle cues (e.g. breathing, sighs, chair movements) explain how mediums could obtain 60 - 90 % accuracy for obtaining specific names, initials, and historical facts - facts that selectively apply to specific individuals?
Is there any scientific evidence documenting that subtle cueing can itself be used to obtain the amount of accurate specific information reported in Schwartz et
al? To the best of our knowledge, there is no scientific evidence that 'cold reading' can generate specific information of this kind
using only subtle cues of this sort (i.e. not using pre-show / pre-experimental information, and not using yes / no verbal feedback).
Sceptics who speculate that 'cold reading' can achieve similar results have a responsibility to show that identical findings can be obtained
under the conditions used in the Schwartz et al. research (e.g. the single-blind sitter-silent condition that effectively rules out pre-experimental information and verbal feedback). We welcome such experiments.
In addition, Schwartz et al. stated in the paper how their current research was employing long-distance readings where
all visual, auditory, and even possible olfactory cues were eliminated. A paper using this procedure is scheduled to appear in the October issue of the
Journal of the Society for Psychical Research (Schwartz & Russek, 2001 b).
Wiseman and O'Keeffe curiously did not mention that we were fully cognizant of such issues and were actively researching them at the time the Schwartz et al. paper was published. The erroneous implication from their critique was that Schwartz et al. were not aware of these historic (and important) considerations.
Possible Judging Bias: Can it Explain the Findings?
purpose the original Schwartz et al. experiments (2001) was not to rule out possible rater bias, but to minimise it. Sitters were selected because of their commitment to research. They endured time-consuming conservative scoring sessions where they rated every item from their respective transcripts. They were requested to justify each answer, and indicate if another living friend or family member could validate their rating. Moreover, the summary statistics were limited to the most definitive '+3' ratings.
Is it plausible to speculate that these particular sitters, who scored their data under a high degree of experimental scrutiny, would forget or confuse the initials or names of their deceased children or parents, or the precise causes of their respective deaths?
Moreover, is there any scientific evidence that rater bias of this kinds operates under conditions similar to the present research (e.g. false memory for the initial and name of one's deceased son's name)? We know of no such data.
Wiseman and O'Keeffe went on to suggest, correctly so, that rater bias could be eliminated using double-blind procedures. However, they failed to mention that Schwartz et al. discussed this kind of experimental design at some length, and described studies in progress employing these procedures. The results of a preliminary double-blind mediumship study are currently under journal review.
Though double-blind studies have the potential to definitively rule out rater bias, this does not mean that the current studies can be explained (or be explained away) as rater bias. Common sense, coupled with available scientific evidence, indicates that memory for personally significant specifics such as initials, names, and causes of death - in appropriate individuals (e.g. research sitters) collected in appropriate circumstances (e.g. conservative laboratory scoring conditions) - can often achieve valid results and should not be dismissed.
"Inappropriate" Control Group: Controlling for What?
purpose of the original Schwartz et al. experiments (2001) was not to include an ideal control group, but rather to address, and possibly rule out (or in) one possible explanation for the data - i.e. simple guessing. The simple-guessing hypothesis was of particular interest concerning the "sitter-silent" condition where no visual or verbal cues were available to the mediums.
Schwartz et al. discussed limitations in the control group sample (e.g. subjects were not age, sex, and education matched with the sitters). However, the control group that was employed was not "inappropriate" for examining the simple guessing hypothesis as Wiseman and O'Keeffe implied; it was a straightforward way for estimating rates of guessing by intelligent individuals. Of course such a control group would be inappropriate for addressing other questions, but that missed the purpose of its inclusion in the original Schwartz et al. experiments in the first place.
Wiseman and O'Keeffe correctly point out that because these research sitters had at least six departed loved ones, that the chances of getting specific information by chance (guessing by the mediums) is correspondingly increased. However, the same logic would apply to the simple-guessing control group. Moreoever, as described below, this hypothetical concern does not explain the details of the data actually obtained in the experiment.
The propensity for Wiseman and O'Keeffe not to share certain information - and to ignore other information - in their critique is consistently evident in their selective reporting of the actual data.
Selective ignoring of important data - what do the data actually
Wiseman and O'Keeffe intimate that much of the readings involved general or vague information that a majority of subjects might endorse as accurate. However, they ignored the fact that in Study 1, for example:
(a) all five mediums reported a deceased son,
(b) three reported the initial M,
(c) one reported the name Michael,
(d) three reported that the death was quick, and
(e) two reported (or implied) suicide by gun (all of this information was correct).
In addition, they also choose to ignore the fact that:
(a) none of the mediums reported a deceased daughter (which would have been an error),
(b) none gave a different initial for the son (which would have been an error),
(c) none gave a different name other than Michael for the son (which would have been an error), and
(d) none suggested that son died from cancer or drugs, etc. (which would have been an error).
Is such information general or vague? Can this pattern of accurate observations be credibly explained as rater bias or subtle cueing? Given their theoretical persuasion, it is understandable (but not justifiable) that Wiseman and O'Keeffe choose to ignore these (and other) data in their one-sided
Even when Wiseman and O'Keeffe cited statements listed from the raw transcripts, they selectively ignored data that did not support their contentions.
For example, they included how one medium in Study 1 essentially said "once, twice, thrice" concerning the loss of three children. However, they selectively did not report that this medium went on to talk about a miscarriage that the sitter had kept secret from her husband (as well as the experimenters who were blind to most of the information pertaining to the sitters in Study 1).
A Call for Balanced Evidenced-Based Scepticism
sceptical critics are allowed to selectively choose which data they wish to address, and ignore those data that are inconsistent with their critical speculations and contentions, they are, in my personal opinion, not practising responsible science. Who is engaged in deceptive science, the mediums or the sceptics (e.g. Schwartz & Russek, 2001a)?
criteria for balanced and responsible scepticism, with integrity, would be
beneficial. If sceptical criticism is to be taken seriously, efforts should be taken to practice balanced evidence-based
There is no question that research addressing controversial questions should be addressed with scepticism, and Wiseman and O'Keeffe's comments are, in principle, appreciated (and regularly considered in this research). However, sceptics have a deep responsibility to 'practise what they preach' and offer their concerns in an honest and balanced fashion.
People often ask me, "Whose side are you on, the mediums' or the sceptics'?" My response is simple. I say, "I am on neither side, I am on the side of the data. If the data take me to the mediums, I go to the mediums; if the data take me to the sceptics, I go to
Our laboratory's approach is simple: If it is real, it will be revealed; if it is fake, we'll catch the mistake. Our motto is 'let the data - all the data - speak'.
Lacking credible balanced criticism, I have come to the following tentative conclusion:
totality of the data are considered, the most parsimonious explanation that accounts for the largest amount of the data is
currently the continuance of consciousness hypothesis (Schwartz, with Simon, in press). Future research will confirm or disconfirm this current summary conclusion.
In a recent letter published in the
Journal of the Society for Psychical Research, I ended it by citing a statement quoted in the Daily Telegraph:
"Wiseman said, "I am not very impressed with what
I have seen so far." If Wiseman's and other sceptics' biased criticisms are to be taken seriously (e.g. see Schwartz & Russek, 2001a)
I should perhaps say to Wiseman: "I am not very impressed with the criticisms I have seen so far."
Schwartz, G. E. R_ Russek, L. G. S., Nelson, L. A., & Barentsen, C. (2001). Accuracy and replicability of anomalous after-death communication across highly skilled mediums. JSPR, 65 (862), 2-15.
Schwartz, G. E. R. & Russek, L. G. S. (2001a). Who is "pretending" - John Edward or James Randi? Journal of Religious and Psychical Research, 24, 110-114.
Schwartz, G. E. & Russek, L. G. (2001b). Evidence of anomalous information retrieval between two mediums: Telepathy, network memory resonance, and continuance of consciousness. JSPR, 65 (864), 257-275.
Schwartz, G. E., with Simon, W. L. (in press). The afterlife experiments. New York: Pocket Books (division of Simon and Schuster).
Wiseman, R. & O'Keeffe, C. (2001). Accuracy and replicability of anomalous after-death communication across highly skilled mediums: A critique. The Paranormal Review. 19: 3-6.