I agree. We tend to universalize our traits: " Leos are generous?.Isn`t everyone". My answer is : No

Ah but your perceptions, and mine, are of course largely irrelevant.

What a critical thinker would say is, for example, we will measure what a hundred Leo?s spend on presents over a 2 year period (as a proportion of their income) and compare this to a randomly selected sample of non Leo?s.

For some astrologers there are all sorts of problems with this as they might argue your average Capricorn is likely to spend more than the average Leo as buying expensive presents increases their 'status' in the eyes of others and so on. The idea here being that although Leo's are more inclined to be generous, are more subjectively generous, feel more generous, etc this may not reliably manifest in their measurable overt behaviour. Thus we get into the realms of non falsifiable claims.

I felt I should perhaps chip in, since I was the interviewer. That interview - also posted on my site at www.astrozero.co.uk/astroscience/research.htm- was conducted for my book Astrology in the Year Zero, where a shorter version of the interview was part of a broad discussion of astrology's nature.

Dennis Elwell objected to the arguments of Dean et al, and a series of exchanges ensued between them which can be found here:

www.astrozero.co.uk/astroscience/elres.htm

Geoffrey Cornelius makes several references to the interview in the second edition of The Moment of Astrology, and more generally presents a very rich and interesting perspective on the whole question of what astrology is.

And there's some more from me on that question in a couple of articles:

www.astrozero.co.uk/articles/Doubt.htm

www.astrozero.co.uk/articles/modern_science.htm

I hope some of that will be of interest!
Garry

Thank you Garry for chipping in and providing us the relevant links.

I had previously missed your Doubt article but I am glad I came across it at last as it has helped my understanding of the question.

Regards

Matt

Thank you Gary for your book and these additional articles, I was unable to open the 'Summary of the Interchange' which was a shame as I was looking forward to reading your views here.

I can't see why if Astrologers maintain one chart ''works'' better than another, which it seems to me even Cornelius does, why this can't be tested seriously. In my view the best, if not the only way, to avoid all these confounding variables - not least the idea that if it's tested it may not work - is to set something up where neither the client or astrologer knows what is going on. This would be quite straightforward to do. Have Dean and Co done this, if so I would love to see how they went about it.

Hi Mike,
Thanks for pointing out that link had died - I've now updated my site with the new one:

www.astrology-and-science.com/o-attk1.htm

As to your question - I think that between them, astrologers and critics of astrology have tried most things in testing astrology over the years. There were the NCGR/Berkeley double-blind tests - see

www.astrodivination.com/moa/ncgrberk.htm

(and, ideally, read this in conjunction with Appendix 2 of the 2nd edition of The Moment of Astrology).

Is this the kind of approach you were thinking of? And if not, would you like to set out a plan for how a test of the kind you have in mind would be set up?
Cheers, Garry

There was a lot of media coverage this week about research carried out by Dr. Trisha Stratford of Sydney's University of Technology. She found signs in the brain and body of a co-ordination between counsellor and questioner at a certain point in the consultation.

This is nothing new and is a familiar experience for astrologers, doctors and the like. This "sixth sense" does not explain away anything but it may suggest why these type of tests "don't work" The brain switches off looking at a lot of charts rather than switching on in a real situation.

Can we have philosophy as well? I came across this in a talk about Thomas Aquinas

"The more keenly we analyze the laws, the workings and the products of human intelligence, the greater reason have we for humbling ourselves and acknowledging our ignorance. And feeling how limited is our own span of truth, we will readily make allowance for those whose opinions we cannot logically endorse" Edward Pace

Thank you for those links, Garry!

I had not read nearly such a detailed report of the Carlson test before, having come across only mocking references in sceptical literature to astrologers having predicted that they would correctly match about 50% when faced with a one-in-three chance, only for chance to have prevailed.

Without having yet read Professor Eysenck's or (I presume) Geoffrey Cornelius's critiques, I had my own qualms about the test design here, while the conclusions drawn by Carlson smack of being a nonsequitur at best.

This is merely a first reaction to point out what to me seem some of the most obvious errors and problems, not an exhaustive statement of the limitations of the test design.

1. The first test, subject selection of own natal interpretation, is only as good as (a) the fullness and accuracy in accordance with astrological principles and time-honoured observations of the assessments by the participating astrologers, whether singly or collectively; (b) the skill of those astrologers in translating those assessments into verbal language; (c) the ability of the individual subjects to interpret that verbal language in the same way and with the same weighting as each astrologer intended it to be read; (d) the accuracy and impartiality of the individual subjects' self-assessment; (e) the degree to which the individuals made their choices based on self-assessment as compared with the degree to which they made them based on other factors possibly including but not limited to (i) second-guessing which of the interpretations they thought were more likely to have been made of their personalities by astrologers; (ii) which style of wording, expression or page layout resonated with them the most irrespective of the underlying meaning of the words; (iii) the chance of order and position in which the three papers were presented to them and its possible correspondence to their subconscious response drivers. With all these potentially confounding variables at play in the chain of communication and the subjects' own cognitive centres and thoughts, the trueness to life of the astrologers' inwardly held assessments of the personalities of the individuals is not fairly being tested.

2. The sample size is potentially a very limiting factor here in view of the many potentially confounding variables, with only 83 subjects being tested, corresponding to expected norms of around 27 and two thirds in each of the results. At this relatively low sample size, a result that appears to be close to chance might be as misleading as one that appears to be skewed several points above or below it. Indeed, it is notable that the control group presented with interpretations of which only the sun sign corresponded to them personally appeared to pick the correct texts for their sun signs significantly more often than those in the test group did, highlighting the limitation of the sample size.

3. The one consistent trend in the results is that in both test and control groups, apparently significantly fewer subjects than would be expected by chance picked the readings intended for them personally as their last choice. This might not be insignificant. While there might have been elements in one of the false texts that frequently spoke to participant subjects on some level that made them think it was more likely than the correct reading to be meant for them (or for their sun sign group, in the case of the control group), it was very much less often the case that both the false texts had such significantly persuasive elements to them than it was that (a) only one or (b) neither of them did.

If we combine the totals for the subjects and control subjects rating the interpretations for them (or for their sun sign) as least fitting to them, we reach a total of 40, as compared with 70 for those thinking it most fitting and 67 thinking it second most-fitting. This is a very big difference indeed.

Surely a larger-scale study is needed to clarify the interpretation of these results.

4. The very weakly positive results of the subsidiary test to Part I in which astrology was not even being tested; only the accordance of CPI profiles to subjects' own CPI self-assessments was, highlights the potentially distortive effects of just one of the several problems I identified in point 1. above, that of verbal communication and its variable cognitive interpretation.

5. In Part II, The CPI profile tests in which astrologers were put to the test matching horoscopes to CPI profiles show a similar trend to those in Part I, with signifcantly fewer astrologers placing the correct profile last as compared with the numbers placing it first or second. Overall the astrologers' blind performance is only slightly poorer than that of the subjects' own abilities to reconcile their own CPI test inputs with the corresponding CPI assessments fed back to them.

Philip

Gary,
I misread this link thinking it was your summary but in fact it's the researchers, useful in this more condensed form.

The Carlson test strikes me as having all sorts of methodological problems. I have read the Cornelius book but it?s not available to me at the moment and I have no memory of his thoughts on this experiment, although I can imagine a few of them.
I found a similar one,
. http://www.scientificexploration.org/jo ... mcgrew.pdf
''After another failure this is some of the conclusion:
In many cases, the correct answer contained the attributes we had chosen, but in
A different [astrological] position. . . . one big mistake was in agreeing to use young
subjects. This was the SaturnINeptune conjunction group, of course, which produced
many 'lost souls' . . . Like medicine, the law, and theology, astrology may not always
give quantifiable results-but it works, nonetheless''. (Mull, 1986).
These remarks do appear to be foolish.

The CPI is unfamiliar to me. I have now read the Wiki entry. In my experience these types of test are too vague to be comparable to serious astrology.It says people from the NCGR were satisfied with the use of the CPI, so I would question the skills of the astrologers who were on this advisory panel. (20% of the astrologers approached agreed).It was very revealing, as Philip points out, that subjects could not locate their own CPI profile from a set of three.
Plus, more crucially, the following:
Astrologers were asked to comment on these areas:
: 1) Personality/Temperament; 2) Relationships; 3) Education; 4) Career/goals; 5) Current situation.
Without access to any of the astrological interpretations I can?t assess their quality. Here we have 5 categories for interpretation. In my view a Natal Chart will give you very little certainty, or even useful information, about 2-5. Time Twins for example, do they show much similarity?

This analysis of tests strikes me as a good overview, includes some thoughts on the Carlson experiment.
http://www.williamjames.com/Folklore/ASTROLOG.htm


So how to do it, well here?s an idea.
Ask 20,000 people to fill in questionnaires, perhaps 500 random questions. Hidden in these would be time, place and date of birth and knowledge of astrology, we can assume an amount of Sun Sign exposure. At the end you may have 5,000 people who were able to provide the required data. Some of these times would be unreliable but with a big enough sample this would become less relevant. (Plus at the end of the experiment work could be undertaken to discover more about the times given).

Then randomly select 400 from the 5,000 and find 40 astrologers. Each subject would be informed they are due to see an astrologer, psychologist, physiotherapist, car mechanic, and piano teacher. Each would be given a sealed envelope to address and post to each of the above, or a pre-written email. Subjects would visit the astrologer on 3 different occasions.

25% of these times would be wrong. Some astrologers will check with the client the birth time at the outset or during the consultation, but as this has already been stated in the letter or via email, that the time is accurate, i.e. from hospital records, mother memory, birth certificate, etc. they may be less inclined to. The experiment would probably need to go on for a couple of years to avoid any astrologers becoming suspicious and information would need to be sent differently each time. Subjects would also be told explicitly that the astrologer has the information required.
Prior to the meetings with the astrologer subjects will be fitted with, without their knowledge, a recorder and camera.

How to assess
1/ - Astrologers should know when the time is wrong this is what they claim to be able to do, albeit more implicitly than explicitly.
2/ - Feedback from subjects should be less encouraging when the wrong time was given.
As the sessions are being monitored this can be assessed more objectively. As you can see from the Carlson attempt Self-assessment is unreliable. Astrologers will be told that certain times are wrong later on. They will be told times are wrong when they are and when they are not. All this will give clues as to the validity of their approach. (Perhaps for 1 person per astrologer the day or year would be out rather than the time, to mix it up a bit).

In order for the experiment to be worthwhile the very best astrologers would need to be involved.

Reasonably rigorous information would become available indicating whether astrologers knew times were suspect. All this would need a lot more thought and much fine tuning, it would cost a fair bit of money (?1 ? 1.5 million approx) and take a good few years to complete, but it would be considerably more worthwhile than the experiments I have seen to date. Dean and Co appear to have debunked a number of astrological claims made by certain astrologers but in the main they seem to have focused on that which seeks to predict real world events, and wishy washy astrology practiced by any Tom, Dick or Harry. This stuff probably should be debunked.

GarryP wrote:Hi Mike,
Thanks for pointing out that link had died - I've now updated my site with the new one:

www.astrology-and-science.com/o-attk1.htm

As to your question - I think that between them, astrologers and critics of astrology have tried most things in testing astrology over the years. There were the NCGR/Berkeley double-blind tests - see

www.astrodivination.com/moa/ncgrberk.htm

(and, ideally, read this in conjunction with Appendix 2 of the 2nd edition of The Moment of Astrology).

Is this the kind of approach you were thinking of? And if not, would you like to set out a plan for how a test of the kind you have in mind would be set up?
Cheers, Garry

http://www.astrology-and-science.com/d-rese2.htm
The Carlson test was diregarded by Smit and Dean. See here

Hi Everyone,
Mike, your proposal for a test is interesting as a thought-experiment. There would surely be ethical issues (e.g. the hidden cameras) and financial issues (the ?1.5M) to be addressed before anything could actually happen though, don't you think?

I wonder if it could be boiled down to simply testing whether or not astrologers can find the right time of birth, through dialogue with test subjects. You'd think tests of that kind would have been conducted by now, though Dean et al at p.469 of Recent Advances say "there have been no adequate blind trials of rectification accuracy". (Though that's 35 years old now so someone may know of more recent tests on this front...?)

Personally, and I suspect I may have said this before, I'm surprised that astrologers haven't shown more interest in the study of time-twins, which seems as if it should cut out many of the problems one runs into in other types of study.

Cheers,
Garry

GarryP wrote:Hi Everyone,
Mike, your proposal for a test is interesting as a thought-experiment. There would surely be ethical issues (e.g. the hidden cameras) and financial issues (the ?1.5M) to be addressed before anything could actually happen though, don't you think?

I wonder if it could be boiled down to simply testing whether or not astrologers can find the right time of birth, through dialogue with test subjects. You'd think tests of that kind would have been conducted by now, though Dean et al at p.469 of Recent Advances say "there have been no adequate blind trials of rectification accuracy".

Personally, and I suspect I may have said this before, I'm surprised that astrologers haven't shown more interest in the study of time-twins, which seems as if it should cut out many of the problems one runs into in other types of study.

Cheers,
Garry

I recently watched a doc about Twins and health on Discovery and there was this case of two identical twin brothers. One lived in Duhram, pursuing a typical unhealthy brit lifestyle. The other moved to New Zealand where he led a very healthy lifestyle with lots of exercise,fruits etc. The twin in the UK had a stroke. Soon after they told the healthy twin to have a check up for genetic reasons. To his surprise he was immediately sent to hospital, as he had signs of an upcoming stroke too.
What this showed is that a healthy lifestyle can only delay the inevitable for a while. Genes speak louder. Many twins look alike but theyre very difderent, one being good at lecturing, another at planning as I recently saw in another show.

GarryP wrote:Hi Everyone,
Mike, your proposal for a test is interesting as a thought-experiment. There would surely be ethical issues (e.g. the hidden cameras) and financial issues (the ?1.5M) to be addressed before anything could actually happen though, don't you think?

I wonder if it could be boiled down to simply testing whether or not astrologers can find the right time of birth, through dialogue with test subjects. You'd think tests of that kind would have been conducted by now, though Dean et al at p.469 of Recent Advances say "there have been no adequate blind trials of rectification accuracy". (Though that's 35 years old now so someone may know of more recent tests on this front...?)

Personally, and I suspect I may have said this before, I'm surprised that astrologers haven't shown more interest in the study of time-twins, which seems as if it should cut out many of the problems one runs into in other types of study.

Cheers,
Garry

The big problem in my view is that astrological results are so vague theres no way the ancients could have known Mars in Aries makes one assertive, Venus in Virgo timid, and so on. if it works it should be clear dont you think?