Chapter 13

The French research
In 1975 I was asked by the French chemical physicist Dr Wolkowski to participate in a discussion on the French radio of paranormal physical phenomena. On the day following the broadcast Dr Wolkowski telephoned that as a result of the programme he had been approached by a Parisian, Jean-Pierre Girard, who appeared to have strong psychokinetic abilities. Dr Wolkowski watched objects move about the table without being touched, and video-records were made of thick metal bars being bent by stroking action.
Metal specimens were sealed inside laboratory glassware tubes, and after being offered to Girard they were returned with the seals unbroken and the specimens bent. Rolf Schonbrot's photograph of these tubes appears in Plate 3.5. Dr Crussard, the chief scientist of the non-ferrous metal company Pechiney-Ugine-Kuhlmann, took up the investigations, using the extensive metallurgical facilities of his laboratories; and many observations of impressive metal-bendings were carried out. Minor bendings of metal in sealed glass tubes also took place during the Pechiney investigations.
Naturally the Pechiney metal bars were identified by engraved markings (I mention this obvious precaution simply because it has been claimed to be untrue in an article in the New Scientist),(29) and many of the aluminium alloy bars were sufficiently thick (8-17 mm diameter) for their 25-cm lengths to be beyond the limit of human strength to deform (see chapter 3).
As a metal deforms it work-hardens, so that the moment (force X distance) necessary for deformation through a certain angle is a smooth function of that angle; the function can be determined experimentally for an alloy of a previously defined composition and treatment. It can also be related to certain other properties of the metal, which are measurable afterwards. Thus one can know from studying a deformed bar of a known alloy the moment that would have been necessary to deform it from a previously undeformed condition. For 25 cm X 1 cm diameter bars of aluminium alloys such as AU2T4, which has been widely used in these experiments, typical moments are in the range 20-50 Nm. The mean of the maximum moments that have been produced normally by men on 25-cm bars is 25 Nm, and by women 15 Nm. Therefore many of these deformations would have required enormous strength to produce, and since the observation and video-recording was often good, one may also state categorically that the manual force used was small (say 1 Nm). The record deformation actually achieved by Girard would have required 75 Nm if produced normally; Geller, who was previously researched by Dr Crussard, once achieved 80 Nm and Julie Knowles has also achieved a bend of this order. I was present at a session at which Girard was filmed(30) in a deformation requiring 23 Nm; the protocol was good, and, as the video-record shows, the manual force was minimal. It is Girard's custom to hold one end of a bar of circular cross-section in his right hand and pass his left hand slowly over the other end for minutes at a time; he then lays the bar down on a flat surface and rests for a short while. After repeated attempts deformation gradually appears; even a small deformation can be observed if the bar is rolled on a flat surface.
Jean-Pierre Girard is little more than thirty years old; he was a rejected child brought up in the west of France in a public institution. When he first noticed his ability to produce psychokinetic effects he did not dare to tell anyone. He works in pharmaceuticals and is also an amateur conjuror; he thought his metal-bending ability might be useful in performances, so he practised in order to be able to perform in front of others. He says he had 'to learn to be an exhibitionist'. When the metal-bending occurs, he claims to feel that it is coming.
Professor Dierkens of the University of Mons has researched on Girard's electroencephalographic output, and on other physiological parameters, during the metal-bending. The characteristic feature of the EEG record is the occurrence of alpha rhythms, of highest amplitude in the parietal electrode, on the right side. Before the observed bending the alpha occurs only in bursts, but in the 'second phase', during the actual bend, there is continuous alpha. The peak frequency is 10 Hz, but there is a subsidiary peak at 9 Hz. Alpha rhythms are characteristic of sleep or of a high degree of relaxation (cerebral rest), but simultaneously the heartbeat of Girard rises to rates as high as 160 per minute. Professor Dierkens writes: (31) 'Girard is well conscious that creating PK is similar to experiencing an orgasm.'
When Girard visited me in London I found that he produced signals on resistive strain gauges which were mounted on an aluminium bar 13 mm in diameter and 30 cm long. Three sensors were used, respectively 3 cm, 8 cm and 13 cm from one end, leaving the other end free for him to hold in his left hand. This is not a very satisfactory procedure, since holding the bar in the hand can cause some deflection of the pen-records; it requires observation and experience to sort out paranormal signals, which are nearly always sharp-fronted. We recorded a great profusion of sharp-fronted signals, obtained with no touch by the right hand, and therefore believed to be paranormal. Most were on the 13 cm strain gauge but some were on the 8 mm strain gauge, not always synchronously with those on the 13 cm strain gauge. There were no signals on the 3 cm strain gauge, which is consistent with the fact that no one has reported a bar bent at the very end by Girard; the bends are always somewhere in the middle.
l mounted a competition in signal-production rate between Girard and some of the children, particularly Julie Knowles and Stephen North. Girard's signals were not larger than those of the children; in fact they were smaller than those of Stephen, but they were in greater profusion - perhaps twenty times as frequent.
When his mood is right, Girard applies himself to the task of bending with some intensity; the induction effects (chapter 17) produced on the observers can be impressive. Numbers of senior French physicists have been invited to Dr Crussard's sessions and asked to hold metal gently in their hands while the bending proceeded. I will not uncover their embarrassment when bends occurred, nor reveal their names, without permission. Sometimes a metal bar would become bent in two distinct places when held at one end by Girard and at the other end by a scientist.
The French research(32) has concentrated on the metallurgical aspects of paranormal metal-bending. This has been a suitable approach, since the Pechiney laboratories are thoroughly familiar with the physical properties and structure of the aluminium alloys they have developed for aircraft such as Concorde and the Mirage fighters. The measurements which are commonly made on each specimen before and after exposure are as follows:
1 All the dimensions were measured, especially the deformation from straightness, and the thickness; for bars of circular cross-section the diameter was carefully studied. Accuracies of about a micron could be obtained.
2 The micro-hardness, which we have seen to be characteristic of the granular and dislocation structure of the metal, was regularly measured at a large number of points; for Vickers hardness, a diamond pyramid is forced into the metal and the diagonal dimensions of the square indentation measured under the microscope.
3 The residual strain profile in the metal was measured, using X-ray diffraction techniques which are a speciality of Dr Bouvaist. The principle of the technique is basically that of Bragg diffraction. There is a linear relationship between the proportional lattice strain (delta d)/d and sine squared psi, where psi is a certain angle measured in the X-ray diffractometer. The distance d defines a separation of planes within the crystal. The technique is used with polycrystalline metals, and is not much affected by grain size.
4 Foil specimens were often taken from the metal, and electron micrographic examination made at various magnifications. The grain boundaries are seen at low magnifications, and at high magnifications it is possible to study the forms of the dislocations and count the loop dislocations when these are seen.
5 The scanning electron microscope was used in the back-diffusion mode to obtain the dislocation density or plastic strain from the width of the channelling patterns or 'Kikuchi lines'.(33) These lines arise from the inelastic scattering of the electrons. Their absorption is different in different regions of the crystal, and their width can be related to the dislocation density below the surface.
6 Electron probe microanalysis was used for obtaining the local composition of alloys. The X-ray spectrum arising from the electron bombardment was analysed not with a spectrometer but with an energy-sensitive solid state probe.
All these measurements show changes when a bar of metal is bent, either normally or paranormally. The differences between the two sets of measurements are such as would require examination by a metallurgist in order to give a full interpretation.
One thing that is clear from these studies is that in paranormal bending the 'elastic component' is largely suppressed (see chapter 11). The dependence of applied stress, sigma, upon strain, epsilon, in a metal is typically of the form of Figure 13.1a. When increasing normal stress is applied the strain increases in such a way that a point travels along the curved graph in the direction of the arrow. When the metal is stressed beyond the yield point and the stress is suddenly relaxed at a point A at the apex of the graph, the metal behaves in such a way that the point moves downwards and to the left, reaching the axis of zero stress at B.
Thus there is a permanent strain or extension of the metal, but it is not as large as was the temporary strain at point A. If we were to plot a graph of the time variation of strain it would in a normal bend have the form of Figure 13.1b (full line). The elastic component contributes temporarily a large proportion of the strain. Ultimately the elastic component ceases to contribute, and it is only the internal stress which holds the metal under its condition of permanent strain.
However, in paranormal metal-bending it seems that the path taken from 0 to B is more direct; in Figure 13.1b a possible path is represented by the broken line. In Figure 13.1b one cannot easily know the path taken, but we have seen in chapter 4 that some of the strain gauge signals are of the same form as the broken line of Figure 13.1b.
The stress that is operative in the no-touch paranormal metal-bending process is apparently an internal stress. The residual internal stresses were found by Dr Bouvaist to be somewhat different after a paranormal bend from those after a normal bend.
The residual internal strain profile is related to the profile of the applied stress. Consider the normal permanent deformation illustrated in Figure 13.2: the applied stress increases as one proceeds outwards from the neutral axis. Between the surface and the broken line planes the stress is so large that the yield point is passed and the strain becomes plastic; by contrast, in the inner region, the strain is entirely elastic. This results in the setting up of reverse strain, so that the strain profile takes the form of the final part of Figure 13.2. The residual strain profile of a normally deformed metal bar is governed by these features. But the residual strains inside a paranormally deformed bar can be different. Dr Bouvaist has measured anomalous residual stress on metal bars exposed to the action of Jean-Pierre Girard.
Figure 13.1 (a) Development of plastic strain epsilon p by application of stress sigma which increases from zero to a point represented by the solid circle A, and is relaxed to zero at solid circle B. (b) The time-dependence of strain developing to its final value epsilon p in a case such as Figure 13.1a.
The most significant findings of the French metallurgists have been the modifications of the physical properties of thick metal bars (150 X 30 X 4.5 mm), produced without any bending by the paranormal action of Jean-Pierre Girard. It is because the residual strain profile is anomalous that no bending takes place. But Girard's 'action' in these cases is probably not very different in other respects from what it is in a paranormal bend. The video-observation and the measurements made on the metal bar leave no doubt that there has been no actual measurable bending. Girard claims to feel in some way that there has been action. I have myself observed one of these events, and am of the opinion that if sufficient measurements had been carried out on other metal specimens exposed to other subjects it would turn out that this type of event was by no means so uncommon as appears at present. Up to the present only four cases of 'anomalous hardening' have been properly documented.
The feature of anomalous hardening events, as reported by Dr Bouvaist and confirmed independently by us in London, may be seen in the representation in Figure 13.3. Although there is no bending, on each face of the metal bar there is locaised increase in the hardness.
Figure 13.2 Development of reverse strain in a metal bar during a downwards bend (which for simplicity is not represented as curvature). During the elastic period (first picture) there is maximum extension (horizontal vector to the right) at the top. There is zero strain on the neutral plane, represented by a horizontal broken line. There is maximum contraction (horizontal vector to the left) at the bottom. Thus the end-points of the elastic strain vectors are represented by a diagonal line. increasing stress produces situations represented in subsequent pictures.
There is also a small locaised decrease, typically of the order of 10 micro metres, in the thickness of the metal bar (~ 4.5 mm). The residual stress increases, but not uniformly, from its original value, ~ -l0 MPa throughout. On one face an increase to +80 MPa has been measured, concurrently with a value of-80 MPa on the other. This profile of residual stress would not be typical of a bend, and in any case there was no bend throughout the entire observed event; nevertheless the hardness increased locally.
Figure 13.3 Hardness data on opposite faces of AU4GT4 aluminium bar handled by Jean-Pierre Girard on his visit to London in November 1977. The data points, before handling, are scattered about the horizontal lines and are not shown individually, except for the three crosses in the centre. The standard deviation of all these points was 1.3 VPN. Closed circles and triangles are data points taken by Dr Bouvaist in Voreppe (weight 3 kg). Open circles and triangles are data points taken by Dr Desvaux in Leatherhead (weight 2.5 kg).
Also significant are the increased densities of dislocation loops studied with the scanning electron microscope. The density in the original specimens was typically 7.4 X 10^13 cm^-3: that of the surface exposed to Girard's hand was found to be 130 X 10^13 cm^-3; the density of the middle was 61 X 10^13 cm^-3. Many of the dislocations are loop dislocations, such as are normally formed only by nuclear radiation.
Recently the anomalous no-bend hardening action, with formation of dislocations, has been confirmed independently in Japan by Sasaki and his colleagues.(34)
In some ways the properties of the exposed strip of metal resemble those of a strip exposed to crushing by a weight of 5 tons. This would produce hardening of the correct order and also a decrease in the thickness. However, such treatment would produce a uniform internal strain profile and a uniform dislocation loop profile. The paranormal action is therefore not similar to the normal application of external force.
The strains which are associated with a paranormal metal-bending event are in general not externally applied; we have seen that there are dynamic strain pulses at the metal, and therefore almost certainly dynamic stress pulses; these would probably be associated with the formation of dislocation loops such as are found in the anomalous hardening events. Dr Bouvaist and I decided to perform experiments jointly to see whether there was a quantitative correlation between paranormal dynamic strain pulses and dislocation loops induced in a normal crystal.
We conducted the first experiment with a crystal of AU4SG (US nomenclature 2014), which was exposed to the action of Stephen North, with six strain gauges attached. In three 90-minute sessions a profusion of paranormal signals was recorded, and I summed the total of the signal strengths at each strain gauge.
It was possible that the dislocation densities at these six points would correlate with these summed totals. But unfortunately the background of dislocations in this crystal was very high, due to precipitation during cooling, and significant differences in dislocation density could not be observed after exposure. There was also a high level of surface porosity. The crystal was of a very tough alloy, so perhaps we aimed too high in this initial experiment. A soft pure aluminium crystal with very few residual dislocations was then exposed; no fewer than ten permanent deformations were recorded and a high density of dislocation loops was observed; since these could have been associated with the bends, analysis for correlation was not attempted. Silicon crystals were then exposed, but many cleavages occurred.
Despite our initial failure to quantify the correlation of dynamic strains with dislocation densities, it still seems that one primary mechanism of paranormal metal-bending is the formation of dislocation loops within the metal crystal. Metal-bending has often been found to be accompanied by dynamic strain pulses; and the dynamic strain pulses are often accompanied by dislocations.
The next step is to consider how a dislocation is formed. The normal method of formation requires some force or atomic movement brought about by action in another part of the metal. So the first dislocation cannot be brought about normally without external force, heat or nuclear radiation being applied. We have searched for macroscopic heat and for nuclear radiation, with negative results.
My own speculation is that, temporarily, a vacancy or group of vacancies might be formed paranormally; such vacancies would propagate as dislocations, by normal physical processes. In later chapters, using quite a different approach, we shall see that the appearance and also the disappearance of macroscopic objects from given locations is known to have taken place in the presence of metal-benders and other psychics; it is known as 'teleportation'. Vacancy formation might be regarded as teleportation on a microscopic scale. However, we do not know that the primary event is not the transfer of energy rather than that of particles.
A possible piece of evidence for vacancy formation comes from the investigations by Rauscher and Hubbard(28) of the surface of a crystal of potassium bichromate K2 Cr2 O7 cleaved paranormally by Uri Geller. When electron micrographs of this fractured surface were compared with those of a normally cleaved crystal grown under the same conditions, it was found that the Geller crystal had a number of 'trench-like' features which appear to be rectangular cavities in the crystal bulk. None of these features was seen in the control crystal. It is tempting to regard these cavities (Plate 13.1) as examples of a sort of gigantic vacancy formation. However, the possibility that inclusions of air were trapped during growth should not be overlooked, and further growings and exposures are now being undertaken. Stephen North has achieved an observed cleavage of one of Professor Rauscher's crystals under the monitoring of two strain gauges.
The French researches are also relevant to the questions discussed in the last chapter. How much of the action is paranormal softening, and how much is paranormal force? Or are they both different aspects of the phenomenon? The conclusions reached about the suppression of the elastic component in paranormal bending are consistent with the idea of a temporary softening. But dislocations in general make metal harder, not softer.
But, additionally, Crussard and Bouvaist have obtained the first quantitative evidence for paranormal permanent softening. No bending took place, but the specimen (similar to the previously described one) showed permanent local softening. The aluminium alloy used was again AU4SG. The mean hardness of the particular specimen was 167 VPN; after exposure to Jean-Pierre Girard two soft regions of more than a centimetre in length appeared; in the first region the hardness tapered gradually with distance to 90 VPN, but in the second the fall to 80 VPN occurred as a sharp boundary. Electron micrographs of the first region showed a spotty appearance typical of precipitation in this alloy. Small regions of a stable structure were precipitated from a matrix of metastable structure. This type of behaviour is typical of the alloy after heating to and cooling from 625°. But no heat was applied or observed.
Plate 13.1 Cavities (C) shown up in electron micrographs of cleaved surface of potassium dichromate single crystal fractured by Uri Geller in Elizabeth Rauscher's laboratory
The appearance under the electron microscope of the second region was unexpected. At low magnification the grains stood out against dark intergrain films, which are characteristic of thin regions of liquid. This appearance is typical of a metal heated to near the melting-point and slowly cooled. It is typical of a metal that has experienced quasi-viscous creep.
Of course there is no question that the real temperature of the entire specimen could have reached these high values. However, on a microscopic scale, violent events are indicated by these micrographs. It is in the regions of grain boundaries, and only in these regions, that we have to postulate the rearrangement of atoms. Very few atoms need be involved; only those which form layers of liquid in the normal process known as 'quasi-viscous creep'. We still suppose that the primary event brought about by the psychic action is the local movement of atoms, or perhaps of energy. We are reminded of the 'Maxwell demon', an imagined agency for removing cool atoms so that a temperature rise could take place.
Another branch of the work undertaken by Dr Crussard and his collaborators has been the study of regions of magnetization (chapter 11) produced apparently paranormally in non-magnetic stainless steels. Austenitic-martensitic transformations have been brought about by Girard. This field of study is highly specialist, and I shall not discuss it in detail; it falls under the heading of structural changes brought about by paranormal means, and will be mentioned again in chapter 16.
One modern metallurgical technique not yet exploited by Dr Crussard or by ourselves is that of acoustic emission. The ultrasound pulses emitted during microfracture events at grain boundaries can be recorded with a modern transducer, transferred to magnetic tape, and subsequently inspected and photographed on an oscilloscope trace. In this way one can distinguish between single and multiple grain boundary fractures, and also distinguish twinning. Dr Ronald Hawke(35) has conducted no-touch experiments of this type with a Californian metal-bender (anonymous); the signals recorded were all recognizable as single grain boundary fractures.
All of these thirteen chapters have been concerned with paranormal action on metal. But there are other psychic physical phenomena than these, and in the course of my investigations I have come across various kinds produced by the metal-benders. I shall devote several chapters to discussing them, before trying to fit the whole picture into the physics background with which eventually we must all come to terms. In the next chapter I shall consider thermal phenomena.

Back to books list.
Back to main index.
Back to Uri Geller's home page.