Skyscript Astrology Forum

Mark,

Thanks again for this detailed overview and clarification of your work.

I'd like to come back to the difference between AstroDatabank's AstroSignatures and multivariate analysis.

AstroSignatures [are] different from multivariate techniques.

I own a copy of ADB so I am very familiar with the way signatures work in that program. As you say, the factors are simply added together, albeit one can assign weight to each factor. Am I correct in understanding that the multivariate technique you are referring to is a technique of optimization? In other words, each factor is varied independently of the others so that the largest effect is found. Is this what you meant?

It is as if we are trying to see bacteria with a magnifying glass when we need to develop a microscope.

This is a good analogy, I quite like it. I do believe, however, that ADB's factor analysis, possibly for the first time ever, does have sufficient power to uncover significant factors at work. I am in the middle of such a project right now, and the results are very encouraging. Please see my other post at http://skyscript.co.uk/forums/viewtopic.php?t=1752 for a bit more detail.

Garry,

I wonder if you (or anyone) could recommend a reasonably accessible text that would give a beginner a handle on the basics and and explain terms like 'multivariate analysis' etc?

I have the fourth (1975) edition of Mendenhall's Introduction to Probability and Statistics, which is a good and readable introductory text for people not already experts on the subject. Currently in its 12th edition, this book now includes a CD featuring computational software and interactive visual tools, and sells for around $100. I picked up my copy in a closing-down second-hand bookstore for $7 (no CD though ). The main focus of the book is on statistical inference, which is the ultimate goal in any statistical study.

Hi Deb,

I?m also reluctant to view the process of astrological analysis as something that can be compared to computer analysis, or that a computer could ever mimic the process, regardless of how advanced and sophisticated the technology becomes.

I'm reminded of what chess-players used to say about their intuition which could never be mimicked by a computer, no matter how powerful they become. Guess what, the strongest chess computer outplays the world champion today. Granted, chess can be reduced to a well-defined set of rules, and the number of possible combinations is finite however large. But what if we find out in a decade that computers are better at judging charts than humans? Mark's program already did that in a limited application, so this is not at all far-fetched.

Nevertheless, you may be right that there is more to judging a chart than simply applying a set of rules, which is where intuition or divination would come in. But I think we have to differentiate between the statistical studies on one hand, and the process of judging charts on the other. Past and current scientific studies, with the exception of the one example above, have not attempted to match the judging process. Their objective was identifying significant factors. In contrast to this, judging charts is applying these factors. So while no 'dialogue with the divine' is necessary for statistical research, judging charts may better be done with a certain mindset or attitude.

Kepler for example, in his Harmony of the World, considers it necessary to first offer a rejection of the Aristotelian concept of the working of the mind and the acquisition of knowledge, so that the philosophical approach of Plato can be restored as a foundation to his theory on aspectual meanings.

A very intuitive mention of the contrast between Plato and Aristotle! While the latter represents the accumulation of materialistic knowledge, the former restores our spiritual connection to the divine. And because we as humans are both material and spiritual, both processes can be found in our earthly experience. Statistical studies will always follow the Aristotelian way, but judging charts will likely need Plato's help.

Ultimately I?m sceptical of the view that the real truth of astrology could ever be discovered by that means...

You may be right. Computers will help strengthen the astrologer's toolbox, but astrologers may need more than simple reason for applying those tools.

Garry,

I was never convinced by the model of the mind being similar in nature to a computer, which Dean et al seem to take as a given in this section. If anyone has any comments on the arguments I'd be interested to hear them.

It would take me several hours to respond in detail to every falsity in their arguments, so I will just stay with the basics here.

1. Figure 5 (the superchart) is a joke. We could produce a similar figure for human anatomy (showing all bones, muscles, veins, etc), and use that as proof for the impossibility of the medical sciences.

2. The comparison of astrology to judging aesthetic quality of artwork is supposed to be objectively valid?

3. They have no idea what they are talking about regarding to chess. Chess-players don't approach the problem as described.

4. They are totally wrong when calculating the possible combinations in a chart.

5. Complexity doesn't preclude the discovery of simple patterns. The assumption is that our ancestors were exact replicas of today's man, which is false. Also, the complexity of life far exceeds anything else, and we are still successful in managing and discovering simple patterns.

6. They talk about the 'dismal Vernon Clark results' here whereas in section 9.6 they quote him as 'his results have fatal problems,' i.e. they cannot be used for any argument. Self-contradiction, Mr. Dean.

7. Skinner's pigeons? Please, give me a break, is this a scientific argument?

In summary, their arguments are disappointingly shallow, self-contradictory, based on false assumptions, factually incorrect, pseudoscientific and seem to be agenda-driven. Need I say more?