Uri Geller - a bibliography - homepage

Technology Review - May/June 1976

Magic and Paraphysics


Martin Gardner

... And now the wretched story is happening all over again, with Uri Geller in the center of a cyclone of irrationalism that is churning over the Western world. Geller is a young, personable Israeli who began his spectacular career by performing what magacians call a "mental act" in Israeli night spots. An American parapsychologist, Andrija Puharich, discovered him, introduced him to Edgar Mitchell, the astronaut who once walked the moon and who now runs his own organization devoted to investigating the paranormal. Mitchell financed Geller's trip to the United States and arranged for him to be tested at the Stanford Research Institute by Harold Puthoff and Russell Targ, two former laser physicists now engaged in full-time psychic research. After a series of poorly designed experiments with Geller, Puthoff and Targ published their favorable findings in Nature.2

Although Puthoff and Targ are personally convinced of Geller's ability to bend metal by PK (psychokinesis) and to perform even more remarkable miracles, their Nature report was limited to Geller's power of ESP (extrasensory perception). His most sensational feat was guessing correctly, eight times in a row, the number on a die that had been shaken in a metal file box by "one of the experimenters". It later turned out that Geller had been allowed to handle the box, and that many prior trial runs had been made. Because the experimenters always shook the box before Geller was permitted to touch it, Geller's handling seemed irrelevant, so it was not mentioned in the Nature report. This seemingly trivial detail gave Geller a splendid chance to obtain information by a technique known to conjurers.3 Had Puthoff and Targ been aware of this technique it would have been easy to take steps to preclude it. The fact that they did not makes the dice experiment worthless.

... A notion of how gullible physicists can be if they have a strong compulsion to believe in paranormal events can be gained by considering a dramatic occasion at Birkbeck College, London, on June 21, 1974. Uri Geller was demonstrating his powers for a small group of physicists. The most distinguished man present was David Bohm, a world-renowned expert on quantum mechanics. Also present were paraphysicists John Hasted, Keith Birkinshaw, Ted Bastin, and Jack Sarfatt (who has since restored his family name, Sarfatti), and psychic researcher Brendan O'Regan, who had arranged the demonstration.

Geller's outstanding achievement was producing a "very strong burst from a Geiger counter tube that he held in his hand. The creation of the burst happend almost simultaneously with Geller's expressed intention to create it .... The creation of the burst was correlated with strong breathing and signs of great physical exertion on Geller's part." I quote from a stiring press release sent out by Sarfatti.6 Geller repeated the Geiger counter bit on the following day for the writer Arthur Koestler and others. "Koestler reported a strong sensation simultaneous with the Geiger tube burst," says Sarfatti, and was "visibly shaken for several minutes." Science-fiction writer Arthur C. Clarke, also there, said it was time for the magicians to "put up or shut up."

My personal professional judgement as a Ph.D. physicist," Sarfatti concludes, "is that Geller demonstrated genuine psyhoenergetic ability at Birkbeck, which is beyond the doubt of any reasonable man, under relatively well controlled and repeatable experimental conditions."

Note the clear implication in Sarfatti's release that having his doctorate in physics made him specially qualified to rule out deception. Well, how would a magician with no Ph.D., but with a knowledge not possessed by Sarfatti, have reacted had he been present? Although I am not much of a performer, magic has been my principal hobby for fifty years. When I read Sarfatti's account, the first thing that occured to me was that Geller could have had a piece of harmless radioactive substance concealed on his person. While twisting around in simulated physical stress he could have simply brought the tube close to his beta source. It could have been in the tip of a shoe, above his knee, in his mouth, behind an ear, under a collar, taped to his chest, in back of his belt. It is not hard to obtain a beta source. A luminous watch dial produces excellent crackling in a Geiger counter. When Philip Morrison once asked Sarfatti if anyone had examined Geller for a beta source, Sarfatti replied that no one had thought of such a possibility, and that it was an "ingenious idea." Magicians find this response hilarious.

Was it a "repeatable" experiment as Sarfatti's release states? Perhaps repeatable in front of Ph.D. paraphysicists, but not in front of knowledgeable magicians.8 Indeed, Geller's methods are both old-fashioned and well known. The interested reader can learn most of them by reading the references cited in notes 2 and 3.

The publication of Geller's methods has had, so far, little effect on the mind-sets of paraphysicists. Their reaction is exactly the same as the reaction of their counterparts around 1900 when confronted with obvious fraud by a physical medium. First, they say, the fact that a magician can duplicate a psychic event does not mean the psychic does it that way. Second, even if he does sometimes do it that way it doesn't mean that he does it that way all the time. Every paraphysicist now concedes that Geller occasionally cheats. After all, is not the poor lad under terrible pressure to produce results, especially on television? How can you blame him for using a little presdidigitation when the psi power is not available? If Geller is caught cheating, as he has been many times, so what? Then he cheated. When nobody catches him, what he does is genuine....

... Before the paraphysicist develops elaborate theories to explain how Geller can bend a spoon, would it not be wise to make sure first that Geller actually can bend a spoon? By PK, that is. Now I do not wish to get into trouble with magician friends by exposing methods used by honest charlatans, but perhaps they will forgive me if I consider in detail Uri's most publicized feat. How does Geller bend a car key?

First, it is important to understand that there is no single method. There are dozens of ways to bend car keys, some of them developed by magicians after Geller made the trick popular, and which Geller never uses because they are too complicated and not adaptable to his casual, impromptu brand of magic. But Geller himself has many ways of bending keys, depending on the circumstances. If he is performing for one person, say a reporter or a Gellerite who has asked for a private demonstration, he will do it one way. If he is in front of a large audience he adopts other procedures. The method he uses depends on who is watching, how many are watching, and how closely they are watching. If he suspects a magician is watching, he wont bend the key at all.

Here is a typical scenario based on the observation of many friends, some of them magicians whom Geller did not know were present and who actually saw his exact "moves." Let's assume Geller is in an office with a group of scientists gathered to witness his awesome powers. Some of them believe Geller has those powers. Others are skeptical but curious. None knows much about magic.

In performing for such an audience Geller has one overwhelming psychological advantage over every magician: he comes on as a psychic. A magician is expected to perform his miracles rapidly and cleanly, without fail, while everyone watches like a hawk to catch the trickery. No magician, when he gets up to perform, dares say, "I'm sorry, ladies and gentlemen. I intended to show you my great trick of floating a burning light bulb across the room, but unfortunately I don't feel like doing it. There are skeptics in the audience. The vibes are unfavorable."

The psychic, on the other hand, is under no obligation to do anything, and Geller plays this role with supurb skill. He begins by saying that he is very nervous, being in such distinguished company, and he doesn't know whether anything will happen or not. All he can do is try. Things are more likely to happen, he says, if everyone wants them to happen. The power he has is not peculiar to himself. Everybody has it. So - if everyone will try their best to make things happen, maybe they will. But don't be disappointed if they don't.

This little speach has the effect of discouraging skeptics from voicing doubts. It also gets Geller off the hook if he finds that conditions do not permit him to do much. Most of all it allows him plenty of time to perform the most trivial of tricks. No magician could possibly get away with taking half an hour to make a key bend, but this often happens with Geller. He will borrow a car key, stroke it, nothing happens. He will put it aside and try later. Again nothing happens. Perhaps on the third or fourth try it will bend.

The reason for this delay is that Geller cannot bend the key until he obtains strong enough misdirection to bend it secretly. The secret bending takes only an instant. Most car keys bend easily, especially if they are long and have a low cut notch. Geller prides himself on his strength (he works out with bar bells, Puharich tells us). If you have strong fingers you can bend most car keys simply by resting the key crosswise on the fingers and pressing firmly with the thumb. Stronger keys require pressing the tip against the side of a table, the table leg, the side of a chair, or whatever firm suface is handiest. In any case, the bending can be done in a split second. Of course it must be done at a moment when no one is looking.

To obtain the neccessary misdirection Geller creates a maximum amount of chaos by moving around the room and going quickly from one experiment to another. Here are some of the ploys he has used to get the needed misdirection.

1. Geller has twice tried to bend a key without success. He tries a third time, letting someone hold the base of the key while he gently strokes it with a finger. Again nothing happens. Geller acts disappointed. Everyone is disappointed. He starts to put the key aside once more. No one is paying much attention because the trick has failed. At that instant someone in the room makes a funny remark. Everyone turns towards him and laughs. It is the moment Geller has been waiting for. His hand drops to the side of the chair while he himself is laughing. Who except a trained magician would be watching his hand at that instant? Geller immediately puts they key aside, carefully placing it in a spot where it is partly concealed so that no one can see the bend. He may not try the key again for another ten minutes.

2. Geller is performing for one person. Both are seated in chairs. The key fails to bend. Perhaps, Geller says, they are sitting too far apart. To move his chair closer, Geller's hands drop to the side of the chair. As he moves the chair, the tip of the key is pressed gainst the chair leg.

3. Geller is in his own appartment entertaining a guest. He sits on a sofa behind his glass-topped coffee table. There seems to be nothing near him he can use for a pressure bend. Who would guess the thick glass of the coffee table will serve him admirably? As soon as a bit of misdirection occurs, and the spectators attention is diverted, the key is bent against the edge of the glass.

4. Geller is entertaining a group of people in an office. They are watching too closely for him to obtain the misdirection he needs. Geller is apologetic. So,etimes it helps the metal bend, he says, if there is a lot of metal nearby. He points across the room and asks, "Is that a metal file cabinet?" If he is in a living room he points to a radiator. Every head turns. In that instant his hand lowers and puts in the work. If the key is weak he bends it in his hand. All he has to do now is hold the key at one end, concealing the bend, walk to the file cabinet, let someone hold an end of the key, then miraculously bend it.

5. On many occasions Geller fins it neccessary to leave the room to obtain strong misdirection. In 1974 when he was performimg for a group of people in Ottawa, a friend of mine in the audience told me that, after many failures to bend a key, Geller asked if there was an elevator in the hallway. The large amounts of metal in the shaft, he said, might help. Geller then dashed in to the hallway, his spectators trailing. Sure enough, in front of the elevator door the key bent.

6. Another one of Geller's favorite excuses for leaving the room is to say that runnung water helps a key bend. He actually used this preposterous excuse on the paraphysicists at Birkbeck College. Let me quote the relevant passage from Sarfatti's ecstatic press release:

"Geller then suceeded in bending several pieces of metal by psychoenergetic action. These objects included the blade of a knife and a key belonging to Bohm. The flow of water from a tap onto the metal seemed to make the bending occur more easily."

To a magician this means that the paraphysicists had been watching too closely. Geller suggested flowing water. Everybody moved to a spot where the key could be held under a tap. In the process of getting there, Geller obtained the needed misdirection. He could have bent the key in his hand, against the side of a doorway as he passed through, or in a dozen other ways. The point is: no one is watching on the way to the sink.

It is important to realize that Geller puts the bend in the key before, sometimes long before, he pretends that the actual bending takes place. Lets suppose he finds a chance to bend the key after a second failure. The key has been put aside, but behind something or partly under something so the bend is not visible. Ten minutes later, when he picks up the key again, he holds it so that only half the key projects from his fingers. Because the visible half is straight, everyone assumes the entire key is straight. Sometimes he rubs the bent key back and forth across a table top. The action and sound strengthen the impression that the key is flat. The key is then given to someone to hold at one end while Geller's fingers surround the bend.

While Geller is gently massaging the key he usually asks if the key is begining to feel warmer. Since it is being handled, it is getting warmer, but most people respond readily to suggestion and imagine that the key is feeling much warmer. Geller continues to rub the key. Slowly he lowers his fingers and allows the bend to come into view. It really looks as if the key is bending at that moment, especially if that is what you are convinced is happening. Geller is a master at creating this illusion. He will see to it that the flat side of the key is toward the audience as he lowers his fingers. Then he will twist the key gradually to bring the bend slowly into view. At the same time he will shout excitedly, "Look! It's starting to bend!" All this combines to create a strong illlusion. Many people will swear later that they saw the key slowly bend, the way a match bends while it is burning.

Sometimes Geller will hand a key, already bent after an earlier "failure" to a spectator. If no one else is present, Geller may ask him to hold the key by its tip but above his head where he can't see it is already bent. geller will then anounce that he is going to attempt something he seldom does. He is going to move across the room, ten feet away, and try to bend the key without even touching it.

The person holding the key naturally assumes it is unbent. Geller, ten feet away, tries hard to kake the key bend. He walks forward, examines the bent key, and acts tremendously disappointed. Nothing has happened! Before the spectator has looked at the key - why should he examine it since the key clearly failed to bend? - Geller is anxious to try once more. The key goes back into the persons hand, the hand is raised. Geller moves twenty feet away. Now he feels the power surging through him! He breathes heavily and seems to be undergoing considerable stress. Yes - he knows the key is bending! "Do you feel it bending?"If the spectator is suggestible he imagines he does. Geller tells him to look at the key. Mirabile dictu! It is bent 30 degrees! To his dying day the spectator will insist that Geller was 20 feet way when he caused the key to bend. Moreover, he will insist, Geller never touched the key. Over and over again, reporters whose keys Geller has bent in this way have written that geller bent their keys without touching them. What they mean is that Geller was not touching the keys at the times they assumed they were bending. The fact that Geller handled the keys many times before the great miracles occured seems totally irrelevant. Indeed, they forgot this entirely.

These are a few of the dodges that Geller uses for just one of his little miracles. I have not mentioned all of his key bending techniques. For example, I have said about how Geller can be secretly aided by his friend, Shipi Shtrang, who is often with him, sometimes disguised as one of the innocent spectators. And Geller has other close friends who occasinally "stooge" for him. The use of stooges, a term magicians use for secret assistants, is a branch of magic in itself. And I have omitted other methods, not using stooges, because they are being employed by magician friends who are now more skillful at key bending than Geller.

magicians are, of course, under the enormous disadvantage of being known as magicians. As a result, they are expected to bend a key under conditions far more stringent than those demanded for Geller. A Gellerite will approach the Amazing Randiwith his fist closed over a key. "You claim you can do anything Geller does," he will say. "Okay, I have my car key inside my fist. Now let's see you make it bend."

"May I inspect the key?" Randi asks.

"You may not," says the angry Gellerite. The key is in my fist. Make it bend without touching it. That's exactly what Geller did when he bent my key."

What can Randi do? He may feebly protest that those were not the conditions under which Geller bent the key, but what Gellerite is going to believe him?

...If Geller has the power to bend metals, why is it neccessary to bend them only under the conditions of a magic performance? If Geller possesses paranormal powers, why do they manifest themselves in such picayune ways as bending a spoon? If he can bend a metal bar by PK, why can't he straighten it again? ...


7. Since I wrote this article, Sarfatti had lunch with James Randi, who fractured a spoon and moved the hands of a watch in a way that Sarfatti found indistinguisable from his observations of Geller. This promted Sarfatti to reverse his opinion and fire off another press release (dated November 19, 1975) which begins: "On the basis of further experience in the art of conjuring I wish to publicly retract my endorsement of Uri Geller's psycho-energetic authenticity." This release appeared as a letter in Science News, December 6, 1975, p. 355. "I do not think," Sarfatti writes, "that Geller can be of any serious interest to scientists who are currently investigating paraphysical phenomena." Sarfatti does not doubt that PK powers exist. He merely doubts now that Geller has them....


The following letter from Harold Puthoff and Russell Targ was printed in Technology Review...

Technology Review - October/November 1976

Letters: Harold Puthoff and Russell Targ

In Martin Gardner's article on Magic and Paraphysics some references were made to the work with Uri Geller at Stanford Research Institute. Unfortunately, Gardner's statements concerning what happened at S.R.I. and what we published are grossly in error. We therefore wish to inform your readers of the facts involved, all of which can be independently verified on the basis of information avaiable in the public domain.

To begin, Gardner states that, "Although Puthoff and Targ are personally convinced of Geller's ability to bend metal by PK (psychokinesis) and to perform even more remarkable miracles, their Nature report was limited to Geller's power of ESP (extrasensory perception)."

Gardner is wrong on both counts. We are in fact not convinced of Geller's ability to bend metal, and our negative findings were reorted in the very Nature article to which Gardner refers (R. Targ and H. Puthoff, vol. 252, No. 5476, p. 604): "It has been widely reported that Geller has demonstrated the abilty to bend metal by paranormal means. Although metal bending by Geller has been observed in our laboratory, we have not been able to combine such observations with adequately controlled experiments to obtain data sufficient to support the paranormal hypothesis."A more detailed statement is found in the S.R.I film, Experiments with Uri Geller, the text of which was released accompanying a March 6, 1973, presentation at a Columbia University physics colloquium. With regard to metal bending, the text states, "One of Geller's main attributes that had been reported to us was that he was able to bend metal .... In the laboratory we did not find him able to do so ... [it] becomes clear in watching this film that simple photo interpretation is insufficient to determine whether the metal is bent by normal or paranormal means. ... It is not clear whether the spoon is being bent because he has extraordinarily strong fingers and good control of micro-manipulatory movements, or whether in fact the spoon 'turns to plastic' in his hands, as he claims."

In discussing our dice-box experiment, Gardner goes on to claim that the reported run of correct guesses as to which die face was uppermost was selected out of a longer run which included "many prior trial runs." That is completely false. The facts are exactly as reported in the nature paper and in the S.R.I. film. The experiment was performed ten times, with Uri passing twice and giving a correct response eight times. These ten trials were the only ten; they were not selected out of a longer run - there were no prior trials nor follow up trials, as Gardner claims.

Gardner's errors appear to be due to his taking at face value the erroneous speculations of Geller's self-appointed debunker, the Amazing Randi.

Letters: Martin Gardner replies

I have enormous admiration for the expertness of Puthoff and Targ in one field - verbal obsfucation. Let me take each of their two counts in turn:

Count 1. When I say that Puthoff and Targ are "personally convinced" of Geller's ability to bend metal, I use the phrase in the ordinary language sense, as when an astronomer says he is personally convinced that quasars are not within our galaxy. It is a probability estimate, as are all scientific beliefs. When Puthoff and Targ deny they are convinced Geller can bend metal, they mean they are not "convinced" because they have not proved such ability in their laboratory. Privately, in letters and in conversation, they have expressed their personal beliefs that Geller has such ability. If Puthoff and Targ wish to make precice their present beliefs about Geller's PK powers, let them give it as a probability estimate: How do they now rate the probability that Geller has PK ability? If they rate it low, it means they have changed their minds.

... Count 2. I did not say in my article that in the dice-box test Puthoff and Targ selected ten guesses out of a longer run. I said that many "prior trial runs" had been made. A trial run, in ordinary language, is a practice run. Yet every time someone points out that practice runs were made with the dice-box, Puthoff and Targ obsfucate by denying that the ten guesses were "selected" from previous trials. They were not so selected.

But that is not the point. The point is that a very large number of practice runs were made during which Geller was allowed to handle the box. This gave him all the time he needed to devise a method of cheating when the final test of ten trials was made. That was all I said and all I meant.

Anyone reading the letter from Puthoff and Targ would assume that there were no practice runs. Without drawing upon private information, I content myself with the following published data:

- In the July, 1973, issue of psychic there are two photographs of geller performing the dice-box test. In the first picture we see Geller ecording his guess of the die's face, the closed box about four inches from his hand, while Targ watches. In the second, we see Geller opening the box to check on his guess. We assume it is the box used in the famous test because we see S.R.I. printed on top. In the obsfuscatory Puthoffian-Targian dialect, this is not a "follow-up trial" because it is not part of the test they reported.

In John Wilhelm's carefully researched book, The Search for Superman, just published by Pocket Books, there is a report on dice-box tests conducted in Geller's motel room. Geller did all the shaking, although Puthoff insists it was vigorous enough to  ensure an hones shake." Commented Targ:"he's just like a kid that he had something that made a lot of noise and he just shook and shhok it." Targ also told Wilhelm that in the famous run of ten Geller was allowed to place his hands on the box in a dowsing fashion.

Targ told Wilhelm that S.R.I. has a good-quality videotape of another dice test in which Geller, five times in a row, correctly wrote down the die's number before the box was shaken. Targ first shook the box, then Geller took it and dumped ou the die. Since magicians familiar with dice cheating know a variety of ways to control a fair die when it is dumped out of a box, how about letting magicians see this valuable videotape? Why keep it top secret? Targ was so impressed by this test that he told Wilhelm he suspects that, even in the run of ten totals that they reported, Geller probably used precognition, not clairvoyance, to guess the number he later "would see when he opened the box." Note: Targ said it was Geller who opened the box!

In the sound track of the S.R.I. film from which Puthoff and Targ quote in their letter, the following occurs: "Here is another double-blind experiment in which a die is placed in a metal file box ... The box is shaken up with neither the experimenter or Geller knowing where the die is or which face is up. This is a live experiment that you see - in this case, Geller guessed that a four was showing but first he passed because he was not confident. You will note he was correct and he was quite pleased to have guessed correctly, but this particular test does not enter into our statistics."

Now Puthoff and Targ have not, so far as I know, revealed whether this single trial, which was recorded on videotape, was part of the test of ten trials. If it was not, then surely it was a practice run. If it was, then presumably the entire dice-box test was recorded on videotape. In the interest of scientific truth, Puthoff and Targ should make this entire videotape available to inspection by magicians. It could then be determined unequivocally whether the theory suggested by James Randi, as to how Geller might have cheated, is a viable one. Come, gentlemen, let us see the entire tape! If we are wrong, we will humbly apologize.

Martin Gardner.

Magic and paraphysics Postscript continued ...

Much has happened since. Taylor has reversed his opinion of metal bending. ... P and T are doing their best to forget about Uri as they devote their energies to research on clairvoyance, or what they prefer to call "remote viewing".... Incidentally, Puthoff recently disclosed to a friend of mine some startling new information about the famous ten shake die test. It did not take place at one time, said P, but was extended over a period of several days!

This answers the third of eleven questions that psychologist Richard Kamman (doing his best to obtain information for The Psychology of the Psychic, Prometheus Books, 1980, a book he wrote with David Marks) sent to Puthoff in 1978. Some of the other questions were:

Which trials were actually videotaped and which other trials were filmed? Which trials wre neither taped nor filmed, and which were both? Were the trials all conducted in one room at SRI, or were some done in other rooms? Were some done in Uri's motel room?

Were the videotapes or films run continuously during each trial, or were they at some time switched off?

Is there any reason why you chose a "pass" trial to include in the SRI film on Geller? Was this one of the two pass trials reported in the Nature article?

Is it possible to learn the time that Geller took on each trial after the die was shaken?

This letter was never answered... We will probably never know exactly what happened during the die test unless Uri someday decides to tell all....

This article contains copyrighted material that has not been specifically authorized by the copyright owner. It is being made available for the general purpose of criticism, comment, news reporting, teaching and/or research. I believe that use of this material is covered under the terms of "fair use". If you wish to use this copyrighted material for purposes other than that provided by law, you must obtain permission from the copyright owner.

Uri Geller - a bibliography - homepage