Robert Todd Carroll
about the newsletter
Skeptic's Dictionary Newsletter
Issue # 18
December 17, 2002
(Past issues posted at http://skepdic.com/news/)
I recently completed a review of the edits made by the folks at John Wiley & Sons, who will be publishing The Skeptic's Dictionary next July. Eight years of abuse from hostile critics of the web site prepared me well for the experience. Imagine reviewing over 1,000 pages of computer generated text and finding only two with no editorial marks or remarks. (This, after being told that for a manuscript this long, it didn't require heavy editing!) Four editors took a turn at correcting, cajoling, and quizzing me. I found the experience very sobering, as well as uplifting. The folks at Wiley know what they are doing and the book will be much the better for their hard work and concern for detail, accuracy, and completeness. They are letting me write my book and are making every effort to make it as good as it can be. Of the several hundred entries I submitted, only one has been cut and it was my decision, although two copy editors wisely recommended it. I agreed with them that WWJD? is a bit over the top compared to my other entries and is unnecessary. (No, Mary, I'm not going to add your suggested entry: WWAH? What would Attila the Hun do?)
Thanks to the copy editors at Wiley, I've now taken an interest in faceology (based on an article by Malcolm Gladwell in the New Yorker), ordered a copy of Banvard's Folly: Thirteen Tales of People Who Didn't Change the World by Paul Collins, and am anxiously awaiting the publication next March of What's Wrong with the Rorschach? (No rest for the wicked, as my mother used to say.) I also filled one of the many gaps in my education: I learned that Edgar Allan Poe wrote several things with the hollow Earth as a motif.
I was making changes in the manuscript up to a few minutes before packaging it up for the post, but I'll have one more chance to correct any errors or flesh out any claims. I'm told that I will receive the proofs for my final edits in about two months. So, expect another quiet period from me during the last two weeks of February.
Just a couple of days before I sent back the manuscript to Wiley, I heard from Dave Rubert, a photographer who generously permitted me to post one of his photos on my Bigfoot page. Dave has given me permission to use another photo of his in the Wiley book. I think when you see it you'll find it as interesting as I do.
My plan is to update the web site eventually, so that it includes the revisions I've made for the Wiley book. When this will happen is anyone's guess. I found it too time consuming to try to update my computer text files, much less my html files, while I was doing the review of the edits.
2) New or revised entries in The Skeptic's Dictionary & Skeptic's Refuge
Since the last newsletter, I've added comments on (1) a South Park episode critical of John Edward, (2) a new theory about the origin of life, and (3) the death of Ray Wallace, notorious Bigfoot hoaxer.
Reader Daniel G. Jennings thinks that the "biggest unreported story in American politics is a blatant attempt by President George W. Bush to use our tax dollars to buy the political support of religious groups...through his faith-based initiative."
I have to admit I haven't been following the news for the past two weeks, but I did see today's headlines and they said something about the CIA being given the go-ahead to assassinate people on our government's hit list. I have to say that this kind of talk disturbs me, but apparently it doesn't disturb any of the leaders of faith-based organizations. My view is that Bush can't win on this one: either he's trying to buy votes from religious groups or he's paying them back for their support in the last election. I did read one news account that said Bush's new policy would allow the federal government to discriminate against non-religious charities. This seems un-American and I hope it isn't true. I don't see why atheists who run a soup kitchen for the homeless shouldn't get government funds if funds are available to religious groups who do the same thing.
4) Books we're reading: Cold reading and Gary Schwartz
I'm reading Ian Rowland's The Full Facts Book of Cold Reading 3rd ed. (2002). While the book is full of facts and gives much advice on how to do a cold reading, it is apparent that it does not take much to explain how cold reading works. It seems that there is only one thing that is absolutely essential for a successful cold reading: the client must find significance in your comments, questions, words, visions, and so on. For example, if I am playing the role of an astrologer, graphologist, palm reader, tarot card reader, personality profiler, fortune teller, medium who gets messages from the dead, or any other similar role and I say something like "Michael is very significant," whether anyone will consider this ruse as a sign of anything will depend entirely on whether the client will give any significance to it, such as "Michael was my father's name. He passed on last summer." If I throw out a string of concepts and names--Michael, father, masculine, football, chest pain, unfinished business, and so on, the success of my reading depends entirely on someone finding a significant connection between some of these items. (Obviously, I'm not going to do very well if I throw out these words in such crude fashion. But if I sprinkle them in a bit of carefully chosen narrative, I know that many clients will be willing and able to connect these bits of data into something meaningful to them.) When I'm done and clients wonder how I knew all these things about them, I can say I don't really know anything about them. What I know is that given the willingness of many people to play the role of client or "sitter," I can count on them to do two things: connect bits of data and find significance in them and credit me with knowing things I don't know.
Cold reading cries out for the kind of socio-cognitive analysis done by Nicholas Spanos and other psychologists on hypnosis and multiple-personality disorder. There is a social context in which the cold reader works and this context entails certain expectations from both the client or sitter and the cold reader (the medium, astrologer, and the like). Just as hypnosis can't work on someone who thinks it is mostly a matter of role-playing, so cold reading can't work on someone who will not play the game, that is, will not connect the data to find something personally significant in it. Rowland (p. 65) gives a telling example of the essence of cold reading in explaining "the Push Statement" (statements designed to be rejected by the client at first). He was demonstrating cold reading in a TV production meeting and used "the shoe and the party" push statement (a narrative about an impression involving a shoe and a party) and the name "Charles." Nobody in the meeting could connect the name with the shoe or the party. Then, ten minutes after the meeting ended, a young woman very excitedly told him that she now remembers a party from her teen years at which she broke her shoe while dancing with Charlie! She was very impressed that Rowland had perceived this detail from her past that even she could hardly remember. Of course, Rowland hadn't perceived any such thing, but no matter.
I could not help but think of Rowland's book as I was driving into work one morning last week and the radio was tuned to "Cool 101.9" (Sacramento). It was my lucky day, for the hosts, Joey and Jackie, had Susan Miller of astrologyzone.com on the line. She hit all four principle themes Rowland mentions--love, money, career, and health--as well as all three minor themes, travel, education, and ambitions. And she was helped along by her hosts. They were able to find some significance in everything Susan said. For example, Susan predicted that next year Jackie would be going to the gym or would want to. Jackie was ecstatic because it's true: she wants to start working out. I listened to about twenty minutes of similar forecasts before Susan mercifully was relieved of her duties. I did learn something from Miller, though: astrology is not the study of destiny; it is the study of cycles. I didn't know that. Furthermore, the world is about to enter a really good cycle. You can read all about it in Miller's book The Year Ahead - 2003. She's even got little Year Ahead books for all the sun signs. The key to the success of such books should be evident by now: as long as there are people who will give significance to her prognostications, she will do well. I predict 2003 will be a good year for Ms. Miller.
I'm also reading the latest issue of the Skeptical Inquirer (January/February 2003), which includes "How Not to Test Mediums - Critiquing the Afterlife Experiments" by Ray Hyman.
While I appreciate the thorough analysis and evaluation Hyman gives to Gary Schwartz's so-called "Afterlife Experiments," I don't think Schwartz deserves as much attention as he is getting. The University of Arizona at Tucson, where Schwartz is a psychology professor, should be embarrassed by this man. Instead of waving his credentials around--he has a Ph.D. from Harvard--he ought to take a refresher course in psychology and find out about such things as apophenia, communal reinforcement, confirmation bias, the Forer effect, pareidolia, selective thinking, self-deception, and wishful thinking. Schwartz seems to know nothing about how the mind works and this badly hampers his attempts to do controlled studies in which he tries to prove there is life after death. For those who do not know, he tests people like alleged clairaudient John Edward to see if the dead can contact them. How does he determine anyone has been contacted by the dead? He asks sitters to find significance in the words, questions, and so on of the medium. The sitter indicates significance by evaluating the "accuracy" of the reading. Schwartz then mixes up a kind of alchemical brew made of part accuracy ratings, part statistical formulas, and part well-placed, self-serving claims such as "The findings were breathtaking....This provided incontrovertible evidence in response to the skeptics' highly implausible argument...that the sitter would be biased in his or her ratings....The skeptics' complaint becomes a completely and convincingly implausible argument...."
Someone should do a psychological analysis of this man and detail how he meets the love of his life or so he thinks at the time, her father dies, she wants to make contact with him, together they start their quest and invent the "Russek paradigm," and eventually they delude themselves into thinking they have scientific proof of life after death. Actually, I just did it and it is probably one sentence too long.
Also in the latest issue of the Skeptical Inquirer is a column by Massimo Pigliucci, a botany professor at the University of Tennessee. It is too bad Pigliucci didn't gear his essay on "causes and correlations" to Hyman's essay. There are some serious problems with designing double-blind controlled tests of causal hypotheses when the hypothetical cause is allegedly supernatural. If I give a blue pill to my control and experimental groups, but put X in the pills given to the experimental group while the control group gets a placebo, and everybody in the X group dies but nobody in the control group dies, I am justified in concluding that X most likely caused the deaths (assuming the size of my groups is adequate and others things being equal). But, if I begin my study by allowing that any difference I may observe between my control and experimental groups at the end of my study could be due to paranormal or supernatural forces or beings, I could never be certain to any degree of probability that any intervention I made in my experimental group was a likely cause of any difference I might observe between my two groups. For all I could ever know, the deaths might have been due to God, Satan, angels, the spirit of Linda Russek's father, aliens with superpowers, ESP, or any of a number of possibilities I can never control for.
The whole concept of a controlled study makes sense only if controls of potential x-factors (variables that might be significant causal factors in whatever it is I'm testing) can be attained by a reasonably diligent investigator. If the potential causes are by definition beyond any human being's control, then a controlled study is out of the question. One can no more do a controlled study on spirits or ESP than one can do a controlled study on galaxy formation. The fact that Schwartz, J. B. Rhine, and thousands of other investigators have tested subjects under rigorously controlled conditions is irrelevant. If you control for cheating, "sensory leakage," and other such things commonly controlled for in ESP experiments, you have not necessarily done a control group study. As long as you allow the possibility of causes that, by their very nature, cannot be controlled, you are not doing a "controlled study" in the usual sense of that term.
Suppose somebody did a study of cold readers where they were able to get exactly the same kind of accuracy as Schwartz gets. If supernatural causes are to be given consideration, then for all we know the success of the cold readers could be due to their being contacted by spirits unbeknownst to them. Likewise, it might have been spirits, not anything in the blue pills, that killed all the members of my experimental group in the hypothetical study mentioned above. Of course, it is always possible that everything in the natural world has a direct supernatural cause, but scientifically designed control group studies do not consider such causes as within their purview.
With all due respect to Massimo Pigliucci, his exposition of John Stuart Mill's views on causality seems a bit shaky. Mill may claim--as Pigliucci says he does-- that "causality simply cannot be demonstrated without experimentation." But Mill did not mean that causality can't be demonstrated except by doing a controlled study, which is what Pigliucci says in his article. "Experimentation" is used by Mill as a synonym for "experience." Mill's point was that we can't know causes a priori, not that we must do controlled experiments to establish causality. We have to have experience in order to find out what causes what. As far as I know, Mill never did a controlled experiment in his life, yet I'm quite sure he made justifiable causal claims on a daily basis. At least, they are justifiable if one follows his arguments about causality in his System of Logic (1843). There he gives five logical methods of finding causes, that is, of finding that something is either a necessary condition, a sufficient condition, or both, for something else to occur. Some of these methods are suitable not just for doing controlled studies or experiments, but are equally suitable for doing prospective and retrospective studies. For example, Mill's "method of agreement" states:
This logical principle can be applied to controlled studies, prospective studies, and to retrospective studies. (A prospective study examines a population that has been exposed to a suspected causal agent and a retrospective study examines a population that has shown the effects of some unknown causal agent. For further information consult a text such as Ronald Giere's Understanding Scientific Reasoning.)
Pigliucci's discussion of Hume on causality is also misleading. While it is certainly true that most historians of philosophy think that Hume made important contributions to our understanding of the concept of causality, they think this way because Hume noted that our concept of causality entails the notion of a necessary connection between the cause and effect but that we never experience this connection. Hume believed that all ideas originate in sense experience. So, how do we get the idea of necessary connection if all we perceive is the constant conjunction of phenomena? Hume's answer was Nature or habit, but that answer is not his important contribution. What was important was that he seems to have identified another chink in the armor of traditional metaphysics. In any case, Hume's analysis of causality is not an important step forward in the history of the concept of causality in science.