UFOgate 

 

A missing Pentacle

The so-called “Pentacle Memorandum” convinced UFO researcher Jacques Vallee that the US government had been toying with the official UFO investigations, and that these were a front for something else… if not something more sinister.

Philip Coppens


In Forbidden Science (1992), Jacques Vallee, who was the inspiration for one of the main characters in Spielberg’s Close Encounters of the Third Kind, reports how in 1967 he found Allen Hynek’s UFO files to be in serious disarray. On Sunday, June 18, 1967, Vallee tried to restore some order in the files and “found a letter which is especially remarkable because of the new light it throws on the key period of the Robertson Panel and of Report #14”. This was the report that was also at the core of Leon Davidson’s enquiries and which made him conclude that the US government were using UFOs as part of a psychological warfare exercise.

Jacques Vallee

The report Vallee found was stamped, in red ink: “SECRET – Security Information” and dated January 9, 1953. Vallee has nicknamed the man who signed it “Pentacle”, arguing it was not up to him to reveal his name. Since, others have named Pentacle. What was never withheld was that the memo was addressed to Miles E. Coll at Wright Patterson Air Force Base, for transmittal to Captain Ruppelt, the government’s lead investigator – as far as the public was concerned – on UFOs.

Vallee read how the opening paragraph established that prior to the top-level 1953 Robertson Panel, somebody had analysed thousands of UFO cases on behalf of the US government. The document noted that the majority of these case reports were found lacking in several aspects, and that the panel should thus ideally be postponed. Failing such postponement, a list had to be created about what the five specialists that would serve on the panel could and should not discuss. To quote Vallee: “the representatives of this top-level research group were against convening the Robertson Panel!” But in the end, they could not stop the formation of the panel, which was chaired by HP Robertson, physicist from California Institute of Technology and would go down into history as “the Robertson Panel”. The other four members were Luis Alvarez, Nobel prize in physics; Lloyd Berkner, space scientist; Sam Goudsmit, nuclear physicist at Brookhaven National Laboratory and Thornton Page, astronomer.

When Vallee read the memorandum, he noticed that there were references to a Project Stork, which Vallee had not come across before. The project seemed to be a key determinant in what the panel could discuss – and what not, i.e. what would be kept away from the panel. By preselecting the evidence, the conclusion the scientists would reach could thus be known in advance. It is a well-established practice that is employed in all government enquiries, but which continues to bedazzle the public, who realise the commission’s conclusions are never in line with the truth. This is largely not the fact of the commission, but of the evidence the commission is presented with. If you do not get to see a smoking gun, you can you comment on it?
Of more interest was that the project Stork team had identified pockets of high UFO activity and recommended that these should be specifically studied. But they also added that many different types of aerial activity should be secretly and purposefully scheduled within the area. To quote Vallee: “what these people were recommending was nothing less than a carefully calibrated and monitored simulation of an entire UFO wave.”

Allen Hynek

Upon reading this memo, Vallee drew the conclusion that the scientists on the panel – and the UFO community as a whole, if not the public as a whole – had been led down a path that had been carefully constructed by people behind the scenes. He wondered what kind of game was truly being played and whether this was just one bad apple or just the tip of the iceberg.
Though just one document, it was a hot potato. Vallee recalled how “Hynek once assured me that if it ever turned out that a secret study had been conducted, the American public would raise an unbelievable stink against the military and Intelligence community. It would be an outrage, he said, an insult to the whole country, not to mention a violation of the most cherished American principles of democracy. There would be an uproar in Congress, editorials in major scientific magazines, immediate demands for sanctions.” The memo clearly showed that a secret study had been conducted… so how would Hynek and the public react?

Before showing it to Hynek, Vallee shared the memo with a colleague, Fred Beckman, who agreed that the document clearly showed that and in what manner the Robertson Panel had been manipulated. Vallee considered this indeed to be the greatest implication of the memo: the document proved that the Robertson panel had been manipulated, which “amounted to a scandal of major proportion in the eyes of any scientist. […] Here is a special meeting of the five most eminent scientists in the land, assembled by the government to discuss a matter of national security. Not only are they not made aware of all the data, but another group has already decided ‘what can and cannot be discussed (Pentacle's own words!)’ when they meet.”
Should Vallee show it to Hynek? He drew the conclusion that Hynek had been “too much timid. In many cases he even believed in the explanations he was producing.” Vallee decided instead to sound Hynek out first. On Monday July 10, 1967, Hynek and Vallee were discussing a new contract that Hynek was supposed to sign; it would continue his relationship as one of the main consultants on the government’s UFO programme. What was remarkable was that the contract was not with the Air Force, but with Dodge Corporation, a subsidiary of McGraw-Hill, a textbook publisher. Vallee noted that the contract had Hynek report into a certain “Sweeney” and that the advice was on whether UFOs represented a danger for the security of the US – which in itself is not a scientific question, which was apparently what Hynek was hired to answer.
Vallee specifically noted that the project Hynek was working for was not called Blue Book – the official Air Force title of the project – but Golden Eagle. Vallee asked whether that name had always been used, and if not, under what name the project was known before. Hynek dropped the bomb when he replied its predecessor was called White Stork. (Note both a bird names and in later years, there would be talk of an entire “Aviary” as being the secret handlers of the UFO community in the 1980s.) He then confirmed that in 1953, he worked for the Battelle Memorial Institute, which had been responsible for Stork. This meant – and confirmed – that Pentacle worked for Battelle too. Vallee then asked whether Hynek knew a Miles E. Coll, to whom the memo was addressed. Hynek did not. He did name Pentacle and seven others whose initials appeared in the secret memo. Hynek confirmed that “they were all administrators or staffers with Project Stork, including the man who sent me on a clandestine survey of astronomers in 1952, to find out discreetly what my colleagues thought of UFOs. ‘Pentacle’ himself was a leader of Stork.” When Vallee enquired how Stork related to the Robertson Panel, Hynek replied “practically nil. Battelle wanted to remain outside all that.” However, that was clearly not the case – and Vallee knew it. Furthermore, it was known that it was Battelle that had written Report #14, a report that was directly linked with the panel. Hynek was either not sufficiently informed or downplayed the role of Battelle. Vallee realised Hynek was not lying; he might just not want to face the overall implication.

Throughout the summer, Vallee battled with the question whether or not he should show Hynek the Pentacle memorandum. In August, Vallee asked Hynek to make some enquiries. Could Hynek have a copy of the Battelle’s card catalogue from Report #14? Hynek “was coldly told that they were no longer in existence. Sweeney pretended to be outraged: ‘It’s a crime, it’s unthinkable…’ But Pentacle is still with Battelle, and he has told Quintanila that the cards had actually been thrown away” is what Vallee wrote in his diary.
On September 8, Vallee decided to give, as a parting gift to Hynek, a framed reproduction of The Lady and the Unicorn. He decided to insert, between the picture and the cardboard of the frame, the Pentacle letter. Once Vallee was back in France, he told Hynek where he could find the memo. In October, Vallee received a letter from Hynek, who wrote that he had gone to Columbus to see Pentacle and his team at Battelle. Rather than a confrontation, he merely told them about the letter and got – unsurprisingly – rebuffed. “I quoted enough from memory to get them very worried. They insisted on the fact that their cards had been destroyed and there was no listing, and all that had taken place with the approval of ATIC.”
But Hynek was upset – worried. In March 1968, Hynek reported that he once again went to see Pentacle, but rather than a one to one meeting, Pentacle had four colleagues with him. When Hynek started reading from his notes, Pentacle snatched the paper out of his hand and said it was an old story and did not return his notes. Vallee: “Why should Pentacle worry so much about a simple letter written fifteen years ago?”
By 1969, it became clear that Hynek would not confront Pentacle, but the people knew that Hynek knew it was a sham, and that he was unhappy about being played – he was, of course, not the only one, but one of those who knew. At the same time, Hynek wrestled with the question whether or not he should go public with the story – which would test his theory that if he did, it would lead to moral outrage with the general public and its elected representatives.
Hynek did eventually talk, though what he said was not what he had found out. Instead, he rather sheepishly argued that a new study should be done. At that moment, the Air Force pushed him aside. Vallee: “First they defused the issue by getting their most vocal opponents to testify before bogus Congressional Hearings; then they selected Ed Condon, a physicist who was about to retire, and he signed his name to a report which was a travesty of science, yet reassured the establishment. They used that report to bring about the liquidation of Hynek’s position, but they were careful not to fire him.” That could, of course, have just been the event that would make Hynek go public.

Though the memo never received the impact Vallee thought it would have from a public point of view , the memo did have an important effect on both Hynek and Vallee; both realised that they had been pawns in a game which they never were able to control, if only because both were too timid to shout “conspiracy” from the rooftops. As late as the 1990s, Vallee still shied away by withholding Pentacle’s name. Vallee concluded that “the discovery of the Pentacle document had a major impact on me. It gave me an uncomfortable insight into the practices of government agencies and the high-powered consultants who serve them. […] It was the main raison for my return to Europe in 1967. It made obvious some unsavoury aspects of scientific policy at the highest level. It provided quite an education for an idealistic young astronomer.” In short, it shattered his innocence.

Shortly after the publication of Vallee’s book in 1992, a document which purported to be the Pentacle Memo came into limited circulation among certain researchers. Dale Goudie confronted Vallee with the alleged Pentacle memo. On June 12, 1993, Vallee agreed that “the document you sent me appears to be genuine. It corresponds to the one I saw.” The document does indeed contain confirmation that Battelle Memorial Institute was working on UFO project(s) at the time of the Robertson Panel (January 1953), and apparently could exercise some amount of control over the handling of the subject matter.
The document was of interest to the entire UFO problem too. UFO researcher Barry Greenwood pointed out that this was a top secret document. It showed that the government treated UFOs as a cover… but that the report also suggested – because of no references to it – that no extra-terrestrial craft had landed, or crashed. Vallee agreed that “the significance of the memo comes, in part, from what it does not say. In particular, it makes no reference to any recovered UFO hardware, at Roswell or elsewhere, or to alien bodies.”
To anyone who still failed to see the importance of the 1955 memorandrum, Vallee added that “the Pentacle proposal goes far beyond anything mentioned before. It daringly states that ‘many different types of aerial activity should be secretly and purposefully scheduled within the area’. It is difficult to be more clear. We are not talking simply about setting up observing stations and cameras. We are talking about large-scale, covert simulation of UFO waves under military control.”

Vallee knew, however, that this gigantic revelation – that the US government had made a policy, which it most likely then executed – that UFO waves were fabricated, would go over both UFO researchers and the public head. It is like finding out that Santa Claus does not exist. He lamented that “I find it odd that a [UFO] group […] should fail to see the significance of the Pentacle Memo, which is an authentic document, when so much time, money and ink have been devoted over the last several years to an in-depth analysis of the MJ-12 papers, which were faked. Perhaps the Pentacle memo only proves that scientific studies of UFOs (and even their classified components) have been manipulated since the fifties. But it also suggests several avenues of research which are vital to the future of this field: why were Pentacle's proposals kept from the panel? Were his plans for a secret simulation of UFO waves implemented? If so, when, where and how? What was discovered as a result? Are these simulations still going on?” They are big questions… which the UFO researchers do not want to hear… and which no-one can answer…

SECRET
SECURITY INFORMATION

G-1579-4

cc: B. D. Thomas
H. C. Cross/A. D. Westerman
L. R. Jackson
W. T. Reid
P. J. Rieppal
V. W. Ellsey/R. J. Lund January 9, 1953
Files
Extra [handwritten]

Mr. Miles E. Coll
Box 9575
Wright-Patterson Air Force Base, Ohio

Attention Capt. Edward J. Ruppelt

Dear Mr. Coll:

This letter concerns a preliminary recommendation to ATIC on future methods of handling the problem of unidentified aerial objects. This recommendation is based on our experience to date in analyzing several thousands of reports on this subject. We regard the recommendation as preliminary because our analysis is not yet complete, and we are not able to document it where we feel it should be supported by facts from the analysis.

We are making this recommendation prematurely because of a CIA-sponsored meeting of a scientific panel, meeting in Washington, D.C., January 14, 15, and 16, 1953, to consider the problem of "flying saucers". The CIA-sponsored meeting is being held subsequent to a meeting of CIA, ATIC, and our representatives held at ATIC on December 12, 1952. At the December 12 meeting our representatives strongly recommended that a scientific panel not be set up until the results of our analysis of the sighting-reports collected by ATIC were available. Since a meeting of the panel is now definitely scheduled we feel that agreement between Project Stork and ATIC should be reached as to what can and what cannot be discussed at the meeting in Washington on January 14-16 concerning our preliminary recommendation to ATIC.

Experience to date on our study of unidentified flying objects shows that there is a distinct lack of reliable data with which to work. Even the best-documented reports are frequently lacking in critical information that makes it impossible to arrive at a possible identification, i.e. even in a well-documented report there is always an element of doubt about the data, either because the observer had no means of getting the required data, or was not prepared to utilize the means at his disposal. Therefore, we recommend that a controlled experiment be set up by which reliable physical data can be obtained. A tentative preliminary plan by which the experiment could be designed and carried out is discussed in the following paragraphs.

Based on our experience so far, it is expected that certain conclusions will be reached as a result of our analysis which will make obvious the need for an effort to obtain reliable data from competent observers using the [... unreadable...] necessary equipment. Until more reliable data are available, no positive answers to the problem will be possible.

 

Mr. Miles E. Coll       -2-        January 9, 1953

We expect that our analysis will show that certain areas in the United States have had an abnormally high number of reported incidents of unidentified flying objects. Assuming that, from our analysis, several definite areas productive of reports can be selected, we recommend that one or two of theses areas be set up as experimental areas. This area, or areas, should have observation posts with complete visual skywatch, with radar and photographic coverage, plus all other instruments necessary or helpful in obtaining positive and reliable data on everything in the air over the area. A very complete record of the weather should also be kept during the time of the experiment. Coverage should be so complete that any object in the air could be tracked, and information as to its altitude, velocity, size, shape, color, time of day, etc. could be recorded. All balloon releases or known balloon paths, aircraft flights, and flights of rockets in the test area should be known to those in charge of the experiment. Many different types of aerial activity should be secretly and purposefully scheduled within the area.

We recognize that this proposed experiment would amount to a large-scale military maneuver, or operation, and that it would require extensive preparation and fine coordination, plus maximum security. Although it would be a major operation, and expensive, there are many extra benefits to be derived besides the data on unidentified aerial objects.

The question of just what would be accomplished by the proposed experiment occurs. Just how could the problem of these unidentified objects be solved? From this test area, during the time of the experiment, it can be assumed that there would be a steady flow of reports from ordinary civilian observers, in addition to those by military or other official observers. It should be possible by such a controlled experiment to prove the identity of all objects reported, or to determine positively that there were objects present of unknown identity. Any hoaxes under a set-up such as this could almost certainly be exposed, perhaps not publicly, but at least to the military.

In addition, by having resulting data from the controlled experiment, reports for the last five years could be re-evaluated, in the light of similar but positive information. This should make possible reasonably certain conclusions concerning the importance of the problem of "flying saucers".

Results of an experiment such as described could assist the Air Force to determine how much attention to pay to future situations when, as in the past summer, there were thousands of sightings reported. In the future, then, the Air Force should be able to make positive statements, reassuring to the public, and to the effect that everything is well under control.

Very truly yours,

[unsigned]

H. C. Cross

HCC:??