M.G.'s first book1 described how he gathered two different samples of eminent French doctors. The first came from a single volume2 and comprised doctors who were members of the prestigious Académie de Médecine between 1820 and 1939. He extracted 576 of them of known birth-data. Then he took the three-volume Dictionnaire National des Contemporains 1936-1939 and again extracted all doctors cited, giving him another 508 individuals (excluding those in the previous volume). Here are the combined results, for Mars, Jupiter and Saturn:
|Physicians||Total||KS SCore||Expect||% Excess|
The second group was less eminent. As Michel remarked, "Most certainly the members of the Académie de Médecine are, as a whole, more famous than the 'notable' doctors."3 This showed up as Saturn appearing stronger for the first group than his second. Why should these have a strong Mars? The doctor as surgeon would then have carried a knife around, and the medical art then had quite a bit to do with bloodletting.
We quote from Geoffrey Dean's Astrology Under Scrutiny, with kind permission:
As Dean rightly observed:
In each case every physician was used whose birth time had been available from registry offices. There was no data selection.4
A rather confusing 18-fold division of the diurnal circle is here used. We're not told much about how MG's first book (L'Influence des Astres, 1955) came to use this.
The first issue of Zeitschrift för Parapsychologie in 1958 had these three articles together:
• Carl Jung: ''Ein Astrologische Experiment,' p.81- 92
• Arno Müller: 'Eine statistiche Untersuchung astrologischer Faktoren bei dauerhaften und geschiedenen Ehen', p.93-101
• Michel Gauquelin: 'Der Einflus der Gestirne und die Statistik' 102-123.
Jung was here reporting his famous marriage-synastry experiment, a propos of his theory of 'synchronicity.' Müller was commenting upon it - his article had nothing to do with Gauquelin's, which followed on. Books and articles about Gauquelin never mention this juxtraposition of Jung and Gauquelin. Surely, this was the very first article by Michel Gauquelin - in German!
Years later, the German psychology professor Arno Müller performed a replication on '1,288 well-known German doctors'5 in 1986, and published an account of this in the NCGR journal.6 One finds no references to this, apart from Michel Gauquelin's brief statement that: 'Overall the findings of the present study [i.e., by Müller] mostly can be understood as a confirmation of the Gauquelin planetary effect.'7
Müller used two source-books for physicians: Biographische Lexikon hervorragender Artze (1962), of which 95% of them were born before 1890, plus also the less eminent Kurschners Deitscher Gelehrtenkalendar (1961) which had generally later births. This gave him a total of 2016 male doctors, of which he found birth-data for 1,288. There was some degree of overlap here with the Gauquelin data published earlier.
His 18-page report published in the Zeitschrift fur Parapsychologie gave graphs of his results but no data, which could be why this study remains unknown. His self-published volume of his researches Astro-forchungs Daten strangely did not allude to it8 nor did an article by him in the Journal of Scientific Exploration (1990) make any allusion to it.9 That would be the only positive-result confirmation of the Gauquelin effect, published in Michel's lifetime.
Two years later, Müller decided to check up on a group of Italian writers to see how the lunar link worked. In the original data-sets for writers in Les Hommes et les Astres Michel had claimed to find a large lunar effect of +31% excess amongst his group of 826 European writers. He also found a big negative-Saturn effect (of -20%) in that group, see Section 8 Thus as one might expect, imaginative writers had shown an excess of the dreamy Moon and a deficit of stern Saturn.
Müller re-checked the names from that same group that had been given in the yearly Italian Who's Who called 'Chi è?' of 1948. He extracted 402 names with reliable birthdata, whereas the Gauquelins some decades earlier had only managed to obtain 192! Françoise commented apologetically:
"our omissions were related to time pressure, shortage of funds for this particular research, language problems in this foreign country, but not to a conscious or unconscious desire to manipulate data."10
What Müller found was11:
|Writers||Total||KS Score||Expected||% Excess||Chi-Sq.||Significance|
Müller reported a negative result! Michel must have been depressed to receive the following 'fake news' from a university psychology professor:
28 July, 1989
Sorry I have no good news for you about the results of my investigation. The hypotheses you sent me with your letter of June 29 (thank you for it) have not been confirmed. A look at the two enclosed tables of result will demonstrate it to you. Of course the subsamples are very small, but there is also no tendency if one combines the groups according to your hypothesis.
He wrongly concluded:
As long as no further positive results are obtained, no relationship between the Moon positions at birth and the future fame of a writer can be considered as proven.12
Did this mean that Professor Arno Müller was incapable of seeing a positive result, from his own data? That would seem to be the case! It could be a case of academic timidity, or a prejudice towards scepticism.
Müller also concluded that 'A significant frequency appears only for Jupiter:'
Higher than average zone placements of the Moon for Italian writers was not confirmed. Instead, I observed an increase in such placements for Jupiter.13
That was his second wrong conclusion, because Jupiter's excess here is insignificant. While admiring the scrupulous care he took in the gathering of data, the checking of each birth certificate etc., our credulity must be strained that so totally wrong a conclusion should have been drawn. This hints at the depth of academic prejudice that the Gauquelins were up against.
His positive Jupiter result - if perchance we wish to consider it, though it was not significant - would be unexpected because Jupiter is extravert. The Gauquelins had found a positive Jupiter-effect for journalists, and these have a very different temperament from imaginative writers. This interesting enigma concerning 20th-century Italian writers could be resolved by the obtaining of more data, easily obtainable from other years of 'Chi è?'
Prof Müller omitted female writers, as he had earlier omitted female physicians: 'because their birth date was politely omitted'14 as he remarked. Let us hope that some Italian investigator will access these, which will make this data-set more interesting.
Müller had obtained a lot more birth-data from this source than had the Gauquelins, and was therefore able to check for any bias. What he found 'eliminates any suspicion of intentional tampering with the data by the Gauquelins., he concluded.'15 Françoise Gauquelin noted, with relief: "This statement is most important for our reputation and for the credibility of our professional results in general."16
There are data for 193 Italian writers up on the CURA site, i.e. those gathered by the Gauquelins, and they score 44 Moon in key sectors, a large 37% excess. Overall, for the 402 Italian writers we saw only a 21% excess. These figures are compatible with the notion of the Gauquelins as having been selective: they don't prove it, but nor do they 'exclude' it. Once again, we don't quite endorse Müller's conclusion.
The Müller data is available here, extracted from his Astro-Forschungs Daten using OCR by S.R.
In 1994 Arno Müller completed his third investigation of the Gauquelin theory, with French physicians' birthdata, and did it conjointly with Professor Suitbert Ertel. They located the follow-up 1972 edition of the same volume of eminent Paris physicians that Michel had used. This new reference-book had a total of 1260 distinguished French doctors, including some of Gauquelin's original data, from which they obtained 915 cases having reliable birth-data.17
Six hundred registration offices in France received letters from him, using the headed notepaper of the 'Institut fur Psychologie der Georg August Universitat, Göttingen', requesting help 'in a study of biorhythms.' Only for those in Michel's original group born in Paris (168) was Müller unable to re-check their birth data: someone would have to live there to do that, he explained. This new total of Academie members had 224 not in MG's original sample.
Müller never published his conclusions from this, except in his privately printed Astro-Data Forschung. Françoise's journal Astro-Psychological Problems was then being produced, in which his earlier allegedly-negative result had been published, would he not have wanted to have this positive result there published? Not until the year 2000 - by which time Müller was 70 years old - did an article by Ertel and Ken Irving allude briefly to some results. Alongside a couple of short paragraph, they included the following rather enchanting graph, averring that it was 'From Müller and Ertel, 1994, n=1,083.'
Ertel and Irving explained that:
The aim of that study was to replicate Gauquelin's 1995-1960 planetary correlations, based on the 1939 edition of the Academy's directory (N=576 members), now using the larger 1972 edition with N=915 members.18
In that case, why did their graph use all 1083 physicians? Arno Müller said:
Most of our analysis will use the 915 cases with birth data obtained independently of Gauquelin. But our eminence trend test will include the 168 cases unique to Gauquelin, total 1083.19
Readers can check the above graph, thanks to Suitbert Ertel kindly supplying the original data of 1083 French physicians.20 That data is the same as given in Arno Müller's volume, Astro-Forschungs-Daten.
That data has here been plotted as for the Müller and Irving graph, in 36 sectors and with the data smoothed with a 3-point moving average, where separate dots are the unsmoothed data: and yes, the graph looks the same! 'Key sectors' are highlighted, of rising and culminating for Mars and Saturn, each comprising 4 out of 36 sectors).
A 25% excess of Saturn appeared in the two Key sectors, and that is a powerful result (Chi-squared =14). A weaker 13% excess appeared for Mars (Chi-squared = 4). Using a twelvefold division instead of 36 sectors gives much the same result.
Here Saturn is stronger - as one would traditionally expect from doctors - while Mars is less strong, whereas in his first study it was the other way round. One would have liked some comment from Müller here, as the only person on Planet Earth to have replicated Gauquelin's work and published the results. That is something we will never hear, from that careful but reticent character.
Arno Müller published ten articles about the Gauquelin work, in the Zeitschrift fur Parapsychologie, over the decade 1986 - '96. Why did he never want to publish an account of this massively significant result he obtained, showing a primal polarity between Saturn and Jupiter, totally confirming traditional archetypes? Are not psychologists supposed to be interested in that type of thing? That must remain alas another one of those 'Astro-psychological problems' that are unlikely to be answered anytime soon.
In each of these investigations, Müller went through the data with great care and yet appeared as timidly reluctant to publish or acknowledge a positive result. How renowned his name would be, had he done so! But, he did twice publish his data in full, and therefore his careful replications have helped to establish the birth of a new science.