Transit - The Astrologers Newsletter

A month or so back I needed to check what the current thinking on Gunter Sachs' Astrology File was. I hope Transit readers might be interested in a brief summary of what I found.

Originally published in German as Die Akte Astrologie (Munich: Goldmann, 1997), Sachs' book was published in English translation by Orion (London) in 1998. The cover blurb claims that the reader will find "Scientific Proof of the Link between Star Signs and Human Behaviour", and for a little while the book found a certain amount of favour with some astrologers, even being nominated for (though not winning) a Spica Award in 1999.

By that point however, a comprehensive demolition of the book's claims to significance had been published in Correlation - Vol.17 No.1 (Northern Summer 1998) pp 41 - 49. Here you will find two long letters from Dr Peter Niehenke and Prof Suitbert Ertel. Niehenke and Ertel raise a number of objections to Sachs' study, including his apparent failure to take note of previous statistical work in astrology, failure to control for artifacts (non-astrological phenomena which can create what seem to be astrological effects), and failure to release information so that further research could be done to ascertain the significance (if any) of his findings1.

This, so far as I was aware, was the last word on the subject. Sachs had not addressed the criticisms raised by Niehenke and Ertel, and in fact in the final chapter of his book had announced his intention of having nothing more to do with astrology. Which hardly seemed like the action of a man who has just discovered it to be amazingly significant.

At some later point, however - it isn't clear when - an apparent validation of Sachs' work appeared on his website: http://www.gunter-sachs.de/astrologie/english/expertiseenglish.html

On this page, Jürgen Chlumsky and Dr Manfred Ehling from the German Statistical Office in Wiesbaden report that they conducted a separate study which validated Sachs' findings. Their comments conclude: "From a statistical point of view The Astrology File has none of the serious failings that its critics have attributed to it."

I contacted Prof Suitbert Ertel to ask whether anything had emerged from this study to change the picture he and Peter Niehenke had described in Correlation. He said he had contacted Dr Ehling in 1998, but had been disappointed by his response. Let me quote part of Prof Ertel's comments here:

"I asked him to send me a copy of the report on their confirmation of Gunter Sachs' results. He said that he did not have Gunter Sachs' permission… A scientist reviewing another author's work does not generally renounce his freedom to inform other scientists about the results of his work."

Ertel went on to remark that there is one new element in the Chlumsky & Ehling article: "They refer to one study that had not yet been done when the book was published. I had suggested such a study in my critical Correlation paper (1998) to which, as far as I remember, I drew Ehling's attention. The authors allegedly did what I suggested, but they do not refer to my suggestion, they nowhere provide any reference to any critical study.

"To my surprise, they report positive results of the suggested critical study that they allegedly did. Again to my surprise, that report is very brief, which is surprising because positive results of this test would give extreme credence to Sachs' claim. My impression is, that these authors are well aware of faults of this particular study (if it has been done at all) and therefore do not want to trumpet out this 'result' thus avoiding to draw too much attention of competent critics on it."

"Aside from this result (which cannot be true, I am sure), this piece on the web… has no scientific value at all, its only purpose seems to be to make uninformed readers believe that Sachs' results are beyond dispute."

And that, so far as I am aware, is the most accurate assessment we have of Sachs' work. We don't know for sure that his findings were insignificant, but where he has provided data so that critical re-analysis has been possible (and there is quite a lot of this in his book), fatal flaws have emerged in the way the data was treated. It is claimed that further analysis of the data revealed further positive results, but since these are not made public it is not possible to have any confidence in the claims. As things stand, we can only assume that there are problems with the work which render it meaningless.

Note:

1. German-speaking readers may also want to visit this link: http://www.skeptiker.de/skeptiker/archiv/1998/3/die_akte_astrologie.html where Herbert Basler, Director of the Institute for Applied Mathematics & Statistics at the University of Würzburg, pursues an independent investigation and arrives (I am told) at the same conclusion as Ertel and Niehenke.