Urban-Lurain's 'Astrology as science'

The work of Mark Urban-Lurain is groundbreaking in astrological research (in my opinion) in that it identifies a crucial reason why statistical tests of astrology usually fail to provide evidence for correlation. Simply stated, his position is this: testing single factors such as birth signs will not yield significant results; a multivariate or multidimensional approach (i.e. looking for signatures = several collaborating factors) will. This is entirely in line with how astrology is practiced, and is also born out by his research on alcoholics. The principle is in accord with John Nelson's findings on the quality of radio reception.

What is curious about his work (which was published in book form in 1984) is that it didn't seem to receive any attention at all. Phillipson's 'Astrology in the year zero' doesn't even reference the author, and his work is only mentioned tangentially by Mike Harding on p.113: "I think there is a program that can detect a propensity for alcoholism within a birthchart. I think this is a possibility." That's it.

Why hasn't anybody tried to replicate his results (which are significant)? Why hasn't this important study received attention? A web search didn't turn up anything on this, and (as far as I'm aware) the only book referencing Urban-Lurain's work is David Cochrane's 'Astrology for the 21st century.' I don't have this book so I don't know how extensively he treats the study. Cochrane does state this, however, in chapter 2 (The search for a starting point in astrology): "Given the failure of most astrological research to correlate astrological factors with behavior or personality characteristics, most serious astrological researchers believe that AstroSignatures must be used instead of single factors. The most promising studies in astrology have involved a combination of many factors in an AstroSignature."

Could somebody please enlighten me on the status of this line of research?

Last edited by aquirata on Mon Jul 10, 2006 7:35 am, edited 2 times in total.

I'd agree that Mark Urban-Lurain's work deserved to feature in 'Year Zero'. Unfortunately I wasn't aware of it at the time I wrote the book.

So far as more work on astrosignatures, the most obvious link must surely be Mark McDonough and his Astrodatabank software. In his studies, astrological signatures (at least, as we know them, Jim) have not really worked. I recently interviewed Mark; let me quote him (with his permission):

Mark McD: .... But of course this raises the question, ?what is Mark?s understanding of where astrological research is, at this point when he?s leaving it??

Q: That?s exactly what I?d like to know.

M McD: Let me give you a little history. I met Lois Rodden about nine years ago. I had, in a previous career, designed databases. I was building an astrology database to test out astrological theories for myself. For instance I?d go to a lecture and hear someone saying things like, ?if the ruler of the 10th house is in the 4th house, you?re going to work at home?. I realised there was a need for a database of astrology to test all these things out. I had the knowledge of computers, the knowledge of astrology, the finances and the business savvy to pull it all together, and so I made a partnership with Lois Rodden to use her data, because she had the biggest, most accurate collection to work with. So we formed Astrodatabank.

We did little research projects like looking for which charts have the most planets in fire; or earth; or air; or water. But before I get into that I need to back up? does an ?AstroSignature? mean something to you?

Q: Yes, I?ve seen the description on your website.

M McD: AstroSignatures are a generalised form of what astrodynes did, and Lois was trained in the Church of Light. So we built a generalised ability to count planets by anything. We build a module into AstroDataBank that could count things like: a point for Sun or Moon, half a point for Pluto, quarter points for planets by earth house as opposed to earth sign. We developed a formula which we have listed on the website for getting emphasis by fire, earth, air and water. It was a beautiful study, and the results conformed to the way we all understand fire, earth, air and water. We were studying the really basic principles behind astrology, and I?d say that study was our strongest positive result.

When we started to study things like alcoholism, we weren?t getting Neptune angular and the kinds of things that you?d expect to find in the chart of an alcoholic. So I then proceeded to take the astrological signature function to the max. I said, we have an understanding of the things that an alcoholic might have but rather than test for what we think is true, let?s have the computer count everything in the chart and tell US what factors show ups most in alcholics.

So I built in the capability ? which just came out in AstroDatabank 4.0 ? where it counts everything; 300,000 factors. And it creates a control group. You can take an alcoholic group, make a control group, and say, ?OK computer, go away and work on this? ? the first time we did this, it took several days. But it counted the 300,000 factors, and reported back which factors showed the greatest difference between the control group and the experimental group.

And I consider that to be the most comprehensive study of astrology that has ever been done, because no human would have gone through 300,000 factors to see what was different between a control group and an experimental group. And the stuff that came back was weird! It was not what you would expect, at all. I went to ISAR?s conference ? guess it was about two years ago ? and I said, ?Hey guys, I am not the Amazing Randi, I am not trying to destroy astrology here. I?m one of the good guys, that want it to work! But it?s not working and I want you guys to explain that to me.?

There weren?t any satisfactory explanations. Nobody had an explanation for why the standard stuff didn?t work. And then of course you can always come up with comments such as, ?Well, you forgot the comet Phaedo? or ?You didn?t do profections?, or ?You should have used the day/night formula?.

My response to that is: you remember the old-fashioned TVs where you could use a dial to change channel, but you could also tune it with the outer ring of the dial? Well, one of my purposes in starting up AstroDatabank was to find out, which were the which factors were the channel-changers, the really big ones, and which were the fine-tuners? If the planetoid Phaedo , hasn?t come down in the literature over all these years of looking at astrological charts, it?s probably in the category of fine-tuner. It?s probably a confirming, rather than a fundamental, factor. So when someone comes to me and says, ?You forgot to put in the 5th harmonic? then I say, ?That may be true ? but are we going to put the 5th harmonic in as a fundamental? And if so, why hasn?t that been coming down to us through history??

So that was very dispiriting. Of course, it was also very dispiriting that my partner Lois died. So the truth is, I never got over the fact that I got these weird results, and there was no way to explain it. But I don?t believe this is the last word. Do I think I disproved astrology? No. I have seen far too many amazing ?coincidences? in astrology to believe that it doesn?t work.

So, if it does work in some way, what way does it work? Where I?m standing right now is, that the model I built ? which is to look at 300,000 factors one at a time ? is what you would call a linear model. It?s sort of a single-factor model. Although you can pull the factors together, it?s still basically single-factor. And astrology is too complex for single factor research ? probably too complex for multi-factor linear models as well.

Gauquelin, for example, wrote about Mars in the day-cycle. That?s the only one factor that has borne up under statistical scrutiny. People have done things like, add night and day to get the fine tuning. Typically with this model, the more factors you add, the clearer it will be. So it should be the case that one factor gives you a fuzzy picture, and then ten factors would give you one that was more in focus. And the purpose of the AstroSignatures is, to use the old rule of three: find three confirming aspects before you declare something to be a behavioural trait?. So what I was doing with the 300,000 factors was, looking to see which ones showed the most difference in the control and experimental groups. It can go both ways, by the way. You can have a ?sobering? factor ? where something is more popular in the control group than the experimental group.

So if you select maybe fifty of the factors which show the strongest disparity across the two groups, it seemed that you should be able to make a strong model for what works in astrology. But it didn?t work; at least, I couldn?t see it. Now it may be that there was too much forest for me to see the trees ? that?s quite a possibility. But my belief is that astrology is a non-linear phenomenon.

[End of interview excerpt]

Another relevant study would be the 'New York Suicide Study' which is written up in Nona Gwynn Press's 'New Insights into Astrology'. Again, this looked at a huge number of factors to try and discern a signature for suicide, and came up with a null result.

The picture emerging from these studies seems to be that either a 'signature' will prove to be a much more complex entity than we might have assumed, or that statistics are somehow incapable of capturing the working of astrology.

Hi Gary,

Thanks for this fascinating interview.

What I don't understand is that he seem to be contradicting his own website, where a number of research projects are published in a positive light. He also says in the interview that:

"Where I?m standing right now is, that the model I built ? which is to look at 300,000 factors one at a time ? is what you would call a linear model. It?s sort of a single-factor model. Although you can pull the factors together, it?s still basically single-factor. And astrology is too complex for single factor research ? probably too complex for multi-factor linear models as well."

Is he talking about a limitation in the software? If so, how is this approach different from Urbain-Lurain's approach?

I agree with your conclusion, although I tend to believe in the complexity argument rather than the incompatibility one.

Regarding MUL's research on alcoholics, perhaps it could be added in the second edition of your book, pulled together with MMD's work on the same topic? :)
Last edited by aquirata on Mon Jul 10, 2006 7:35 am, edited 1 time in total.

Hi Peter,

Well, Mark McDonough resigned his position with Astrodatabank some time ago in order to pursue other interests.

I'm not sure if there ever will be a second edition of 'Year Zero', but thanks for the thought!

Re. what Mark means, it's probably best if I let him speak for himself. This follows right on from the previous excerpt from Mark's interview:

Q: Could you explain what that means, please?

A: In algebra you might have an equation such as 3x + 2y + z = the final result. If some of those variables are non-linear, then they should be squared, or to the one half power. That would mean that some variables would interact more complexly than you might think. If you have a ton of oxygen and just a little bit of hydrogen, you?re not going to make a lot of water. There?s a proper ratio where, all of a sudden, bang! ? you?ve got water.

So to find that proper ratio would require statistical tools that are beyond what we currently use. Recall that statistics is a very young science. It only became recognized as a distinct field of study in the early 1900?s. That?s also when basic methods like the chi square and t-test were developed to allow us to be able to distinguish which differences between a control group and experimental group are most likely to be ?statistically significant? or just random variations.

If you look under the formulas for multi-variant statistics it?s a very simple, almost simplistic, formula where you?re saying, ?What?s the average? What does the curve off the average look like? And where is this data point on that curve?? That?s all it does. It?s my belief that, as we do genetic research, that the mathematics of pattern-recognition will be pushed to another level. And when those things become available in the public domain, and not just in someone?s laboratory, then we will have something that might be strong enough to resolve astrology.

The analogy here would be that you could use a set of binoculars to look at Pluto ? and you won?t find it. Use a powerful enough telescope; then you?ll see it. Similarly, computers ? instead of trying to resolve far-away objects, try to resolve complex objects, working (if you like) as pattern-recognition machines.

The net of this is to say that our pattern-recognition software is far too simplistic for the problem called astrology. Even if you?re using multi-variant statistics, it?s still very simplistic. Now, I was speaking with Bernadette Brady about this problem: the archetypes are very wide. So for instance there are twenty different meanings for Mars in Aries; how do you resolve which particular meaning applies for a given person?

My sense is that today?s computer software can?t do it. At this point, without better pattern-recognition, it requires intuition to know which of the seven key-words you should pull out of your satchel at any given time. I believe that astrologers have the experience that every time they read a chart, what they do is fairly straightforward; but what they are doing is a lot more complex than they think. What they are doing in their unconscious is trying to figure out which of the phrases to use. And they won?t use the same phrase for Mars in Aries every time. They?re not like a computerised report generator. And how they nuance those things comes from their intuition and not from simply pulling descriptions from off the page.

So, Bernadette said that when an astrologers reads a chart it starts off as a bunch of dots on the page, and then it resolves itself into a pattern. It?s like the way an optical illusion that starts out looking one way and then changes ? or an even better analogy would be those holographic images, which just look like a bunch of dots, then all of a sudden you see the hidden picture.

Something is going on in that, which I believe is the unconscious doing all of the filtering and pattern-recognition, which is below the level of awareness. In the field of artificial intelligence computing known as ?expert systems? this is a known problem; the classic example is the doctor. When someone?s learning to be a doctor, at the beginning they can tell you exactly all the factors that went into their decision. As they get better at it (?better? meaning not just faster, but more accurate too) they are less able to tell you why they came to the diagnosis that they came to. Sometimes they feel as though they?re making it up. And if they have to explain what they did, they?ll go back to their textbook training and tell you that?s what they saw. But is is post hoc reconstruction, not the real process by which they came to a diagnosis. Again, we have an expert who is doing a multi-variant combinatorial non-linear pattern recognition. Underneath their awareness they are saying, ?Well, if this thing isn?t here, then I can ignore these five other things over here??. .

The software we have at the moment doesn?t do that. It just counts; just adds or subtracts from a score depending on the presence or non-presence of factors. But they don?t combine. And once you start to get into all the combinations, the 300,000 factors that I looked at would look ridiculously small!

Q: So there?s something going on in astrology, you would say, which could be accounted for by saying that it?s super-rational, or that it?s non-rational, I guess.

A: At the moment it absolutely requires intuition to resolve the pattern. Whether at some point, with better pattern-recognition software, we can decrease the amount of intuition ? that?s the hope. And as a matter of fact the analogy here would be Bernadette Brady?s ?the eagle and the lark? analogy. My hope is that, with AstroDatabank, people will be able to start homing in on which are the major factors and which are the fine-tuners, so that we get a more solid sense of what our fundamentals are.

Q: If I could put my skeptic hat on for a moment here, I imagine the response they would come back with to what you?re saying would be, ?If astrology is this skill which people can acquire ? it might be very difficult to explain what you actually do, but all the same people can do it ? then you could cut out the middle-man and just test astrologers; test the ability of astrologers to read charts?. Would you accept that as a valid line of argument?

A: Yes, this is a valid line of argument. The difficulty of course would be, how would you grade astrologers in the first place so you could be sure that, when you did your tests, you were working with the best?

Another point on that matter: apparently Mark Urban-Luhrain did such a test with several astrologers versus his multi-variant alcoholic study, and his computer program was able to predict the alcoholics better than could the real-life astrologers.

But then this brings us to another point again: let?s say you?ve got a great astrologer for your test. Are they a good astrologer at picking out alcoholics? They may be good at telling you things like, ?When is a good time to go looking for a spouse?? If you think about it, what are astrologers being trained by their clients to know how to do? They are being trained by their clients to be able to describe the personality of the person in front of them. And through years of getting signals ? yes/no/not exactly, more like this ? they get better at it. Personality traits are not something which, I believe, a skeptic would tend to put into a test. Personality traits don?t make the kind of definitive tests that skeptics like. So it would be kind of like taking a race-car driver and saying, ?OK, now we?re going to prove how good race-car drivers are at changing tyres?. So there?s an assumption, ?Of course they should be good at changing tyres, it?s to do with a car?. But it?s not something they do every day, so they may be no better at changing tyres than anybody else.

Similarly, if an astrologer is tested on something which is not part of their daily work, then it?s not a good test.

One last thing: I don?t think we?d be complete if we didn?t talk about Geoffrey Cornelius and his point of view ? which in caricature suggests that astrology is like reading tea-leaves. At the extreme, if astrology is all divination, the logical extension of that is that it doesn?t matter what chart you use, at all. And I would reject that argument ? I think most astrologers would reject it as too extreme. ?If it?s all divination, I could do just as good a reading on you if I used Hitler?s chart as if I used your real chart.' That?s not astrologers? experience ? though astrologers do occasionally have great readings with the wrong chart.

Q: Do you think that, when it happens, the wrong-chart phenomenon is valid, that it expresses something about astrology rather than just being a random and meaningless quirk?

A: I think astrology is at least 50% divination. I believe that the archetypes are so wide, that the only way you can deal with it is to have strong divination abilities. So I want to put myself firmly between the camps, and say that astrology definitely involves divination. The astrology we are taught in the text-books, if you took it literally, won?t work. But there are lots of things that do work. And my fervent hope is that, in the next twenty years, people will be using AstroDatabank to get a rough idea of the things that are working and not working by having easy access to lots of good data and control groups for testing things.

As I said before statistics hasn?t changed much since the early 1900?s Or to speak more precisely, the statistics we learn in university is pretty much all vintage 1900-1930. There have been extensions to the basic theory, but little in the way of breakthrough concepts ? at least in standard textbooks.

There is lots of basic research going on in the field of ?data mining.? People are using fast computers and large datasets to look for patterns in stocks, consumer behaviour and genetic information. Right now much of this work is proprietary because it conveys commercial advantage.

And maybe in twenty years? time - when it?s as cheap to get a computer intensive pattern-recognition software as it is to get the standard multi-variant stat packages today ? I will be funding some graduate student who comes along and says, ?OK, let?s send the astrological data through a genome pattern recognition machine?. In the meantime, we?ll keep collecting more data and researching astrology basics. We probably won?t get statistically significant results, but there is lots of work to do to sort through all the astrological factors and get a firmer grip on which ones are the fundamentals and which are the fine tuners.


Reply from Mark Urban-Lurain

Thanks to both Peter and Garry for bringing this forum and thread to my attention. It took a while for me to get an account on this forum, but I?ll try to answer some of the questions about my work.

First, a little background. I did this work for my Master?s degree thesis in the early 1980?s. The focus of my Master?s program was on using statistics and computer models of psychological processes; this research grew out of that. This was before microcomputers (and long before they had the necessary power to do these types of analyses) and this work was done on a large, expensive to access, mainframe computer. Once I had completed the degree, I no longer had ?free? access to the necessary computing facilities, so I was unable to conduct these types of analyses for quite some time.

Second, when I watched good astrologers work, it seemed to me that they are doing pattern recognition, which is something humans are very good at, but computers are much less capable of. For example, humans can quickly identify faces, but even current work in computer science is only beginning to be able to do this; in the early 1980?s there was no work along these lines. However, I thought that using statistical pattern recognition techniques could be fruitful in astrological research. There are a variety of multivariate statistical techniques that are used for pattern recognition work.

For my Master?s degree, my research question was ?are multivariate statistical models better able to emulate the process by which astrologers interpret horoscopes than univariate models?? While I evaluated the models using data of alcoholics anonymous members and a control group, the point of the research was to compare the multivariate techniques (that look at multiple variables concurrently) with the more commonly used univariate techniques (that look at one variable at a time.) A common misunderstanding of my results is that I ?found? signatures for alcoholism. This is NOT true. The multivariate tests were much better at discriminating between the AA members and the control than were the univariate tests. However, the study did not identify ?alcoholism? factors that can be generalized. The statistical reasons for this are beyond this conversation.

Third, the question of why isn?t this study better known? Shortly after I completed my degree, Gary Duncan urged the American Federation of Astrologers to publish it, which they did. However, it was an academic thesis and reads as such. The intended audience was academics who have a background in statistics, not astrologers, who generally do not. I?ve spoken about this work at dozens of US and some international astrological conferences and quickly realized that practicing astrologers do not understand the statistical issues involved. (Actually, most academics do not have substantial statistics training and do not understand the issues.)

In 1983, Geoffrey Dean sponsored an ?astrological superprize? competition to provide ?evidence that the accuracy of chart interpretations cannot be explained by non-astrological factors.? The contest specified that ?the charts must be of ordinary people typical of those who visit astrologer.? I submitted this work and, after almost two years, it was judged/reviewed by Hans Eysenck, David Nias and Ivan Kelly, academics that were certainly qualified to understand it. The judges all agreed that the study showed the power of multivariate analyses over univariate techniques. However, they felt that AA members were not typical of those who visit astrologers. No ?grand prize? was awarded in this competition, but I did receive one of the awards that were given for ?excellence of design.?

Fourth, the question of replication of this study has come up and is appropriate. Science progresses by replication and positive results do need to be replicated, hopefully by someone other than the original researcher. I can only speculate as to why no other researchers have used these techniques. My guess is that it is because these techniques require a substantial amount of graduate level statistical training to understand, let alone to apply. Most astrologers are interested in seeing clients, not in obscure theoretical research. Most academics who are trained in the statistics are not interested in astrological questions or do not understand the complex issues around studying astrology statistically. Hence, the intersection of these two sets is pretty much null.

Meanwhile, I continued to be interested in refining these techniques. However, it was not until the mid 1990?s that personal computers began to have the power and statistical software to do this type of work. Starting then, I concentrated on developing better ways of creating control groups using techniques known as Monte Carlo simulations. My reason is that the criticisms of astrological research generally focus on 1) data sources and accuracy, 2) sampling issues and 3) calculations of expected frequencies (e.g., how often should we expect to find Mars in a Gauquelin plus zone just by chance?) Solid control procedures are needed to address #3 and the Monte Carlo techniques seem to work well. I was pleased that Mark McDonough incorporated a simplified variant of these techniques into AstroDatabank to generate control groups.

I am also concerned about possible violations of the linear assumptions of many of these statistical techniques when applied to astrological research. In my original work I used quadratic versions of these techniques, based on the differing variances of the two groups. My current research investigates some of the more obscure branches of circular statistics to develop extensions of the linear techniques that may provide better fitting models.

A brief tangent to contrast the work I do with Mark McDonough?s AstroSignature work. I am very familiar with Mark?s work. While he does combine a multitude of factors in the various AstroSignatures, the way in which they are combined is additive based on counts of the various factors. This is different from multivariate techniques. In multivariate analyses, the weight of each variable is determined by holding all other variables constant, which I believe is much closer to the synthesis an expert astrologer does when interpreting a chart.

In conclusion, my underlying assumption is that one problem with statistical studies of astrology is that they use very weak (univariate) techniques. It is as if we are trying to see bacteria with a magnifying glass when we need to develop a microscope. The key is not to keep using the same magnifying glass (univariate statistics) and declaring that there is no such thing as bacteria (astrology.) We have to develop better instrumentation before we can hope to see the ?bacteria? clearly. Only then is it reasonable to discuss the nature of these ?bacteria.?

Mark Urban-Lurain

Thanks, Mark, for this interesting response.

Just a couple of follow-up thoughts:

- What you say about astrologers lacking any understanding of the subtleties of statistics is certainly true. 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 think that in an ideal world, one of the astrological organisations would employ a tame statistician who could comment on the validity or not of any scientific/statistical arguments that astrologers put forward.

- There is some discussion of the problem of astrological complexity and the human mind's capacity to juggle complex information, in the interview I did with Dean, Ertel, Kelly, Mather & Smit. It's in Section 17 (begins p.52) of the pdf which can be downloaded here:


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.

I would also like to say thanks to Mark for his very interesting and informative post. It really makes my mind boggle and in a sense I don?t feel qualified to comment on this at all. But it does make you wonder about what we can or might be able to achieve with the help of advancing computer technology. Like you Garry, 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. This seems to me a rejection of the traditional view of astrology as a ?dialogue with the divine?. 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.

Plato?s view was that the human mind is in itself thoroughly informed and that when it seems to learn, it is only really being reminded of what it already knows. Hence things like the meaning of mathematical principles are already determined and we could use inspiration or a heightened state of self-awareness to uncover them or draw meaning form them. Aristotle taught the opposite - that the mind is born empty, like a blank sheet of paper, so that nothing is written on it, but anything could be written on it, providing it is developed sequentially.

Aristotle would probably believe that a computer could be taught the best of human knowledge and with enough sophistication of the technology, there?s no limits to what we can achieve. But Plato would probably believe that *any* human mind would come closer to understanding the symbolic meaning of divine principles than all the computers in the world, because all knowledge is accessible if the human soul is able to make an inspirational leap. I might be with Aristotle as to the best way to develop some sciences, but when it comes to astrology I?m certainly with Plato.

That?s not to say that I don?t support the concept of computer aided research, and I think that what Mark has done is tremendous; I do believe that there are important objective correlations still to be discovered. But ultimately I?m sceptical of the view that the real truth of astrology could ever be discovered by that means, and as Mark shows, even if it could, at the moment it doesn?t seem to be worth holding our breath.

Thanks again Mark, very interesting.

Hi Deb,

I was interested to see this on the forum. I actually have a copy of UL's thesis after studying it a few years ago, and remember feeling extremely frustrated at the time because 1) I had neither the training nor resources to use his model to continue studying astrology with it (I remember feeling like my brains were frying just attempting to understand advanced statistics! lol) and 2) No one else seemed to be interested in it at all! I finally got so frustrated I just had to drop thinking about this sort of study of astrology, and go back to my own empirical observations.

Though I won't be "holding my breath" waiting for someone to "prove" astrology works, I do still have tremendous hope that Mark's multivariate model will be used one day to further astrological research. I have a 12-year-old daughter, and she and her friends sometimes amaze me with their thinking and willingness to look at things in a new light. It's partly their age, of course, but who knows where these Pluto-SCO and Pluto-SAG kids are going to take us?

On another note, I noticed that on the alcoholic astrosignature study, modern rulers were used. Did they ever try traditional rulers, I wonder? I know that Bernadette Brady didn't get significant results with a study she did on heredity until she used the traditional rulers. The longer I study astrology, the more I've seen that the traditional rulers really are the foundation, and tossing them over in favor of modern rulers leads to errors not just on horary charts, but on all sorts of other things.

PS: I'm with Plato. ;-)


There has been a lot of talk at my university lately about making statistics compulsory for all first year students. At first this seemed a bit over the top to me. There are few courses these days where some statistics is not required. All of the sciences, law, business and psychology courses have stats as a compulsory part of the degree. The only ones who escape it are those doing the Bachelor of Arts in areas such as history or English. But I was having a conversation with some students and we couldn't come up with anything where statistics would be of no use whatsoever. Subjects such as philosophy, sociology and linguistics could all benefit from statistics. I did three years of statistics in my first undergrad degree. Unfortunately, that was more than twenty years ago and I have forgotten a large part of it but, even now, it is still useful. I think the multivariate work that Mark has described is definitely interesting and could tell us a lot about astrology.

However, I do not understand a need by astrologers in general to ?prove? that astrology works. Works how? In what way? Under what circumstances? I also do not understand, as I said in another thread, why we think we can take astrology in isolation and expect to fit it to a model that can be proved or disproved. Why is it that so many of us eagerly await scientific or statistical proof of astrology? Deb mentioned Plato and Aristotle. Whether we agree with their philosophy or not, Plato and Aristotle contributed enormously to our philosophical understanding that makes astrology possible. This contribution is not open to scientific or statistical scrutiny but is a vital piece of the puzzle. Aristotle, for example, believed that the south pole was at the top of the world and not at the bottom due to the direction of the stars. His reasons for this appear quite logical when taken at face value and can contribute to our understanding of the worldview that led to astrology. This made complete sense to him. Yet we know that, scientifically, he was incorrect. Do we disregard his argument in its entirety because it does not stand up to scientific scrutiny or do we look at the philosophical contributions he made to the wider concept of the worldview? Nils Bohr said of physics that it only tells us what we can know of the world, not the way the world is. I think this is also true of astrology. Much of what we know to be true today will not be true in 200 or even 20 years time. In the same way, some of what we knew 200 years ago is no longer true today. Carl Jung said something along the lines that the criterion of what is possible in any age is derived from that age?s rationalistic assumptions. What is possible is governed, not by whether it is possible or not, but by our beliefs about whether it is possible or not.

As far as trying to determine whether astrology is able to be scientifically or statistically validated, Jung believed that experimentation immediately creates an artificially restricted paradigm that inhibits an unrestricted answer. It forces nature to give an answer to a question devised by humans. This is very similar to something Dennis Elwell said about expecting human answers to human questions but being met with a cosmic answer we do not understand. Fran?ois Trojani said, that for all its great usefulness, science is a very limited, very fragmented, and not very profound way of trying to investigate the mysteries of the universe. By limiting the boundaries of the question with science or statistics, we limit the boundaries of the answer to one based on science and statistics. How does the cosmic answer fit in here, even if we are listening?

I would also like to thank Garry for making the rest of the McDonough interview available here and Mark for posting a detailed overview of his work. I will reply to these in due course, hopefully within the week. Let me, however, first address Sue's concerns about scientific validation of astrology.

The reason why this is badly needed is as follows:

1. The scientific community considers astrology a pseudoscience, and it makes sure this view is predominant.
2. Because of this, scientific journals generally do not accept astrological material for publication.
3. Also due to this, universities are limited in their ability to allocate funds for astrological research because they themselves cannot apply for funding easily.
4. Furthermore, it is very difficult to set up astrological faculties at universities as a result of this bias.
5. The overall effect is that astrology is being practiced and researched by and within a limited circle of people on limited funding with access to limited resources.
6. The fact that astrological research is almost solely published in astrological or other fringe journals is justification for withholding funding for further research.
7. The fact that graduate-level astrological studies are very limited is justification for withholding funding for astrological research and expanding astrological activities within universities.

It is a vicious circle.

Astrologers can certainly go on practicing as they have done for millennia. However, to truly advance this field of study, one needs funding and universal access to resources. The only way I can see this happen if astrology takes the challenge of modern science.

There is no question that some of astrology can be validated using scientific methods as this already has been evidenced by the Mars effect saga. Yet even the Mars effect has not been admitted to the scientific body of knowledge, so much more research is needed along these lines until the scientific establishment will finally 'get it.'

From the perspective of the scientific community, this is a political battle because they will naturally want to hold on to as much of the research pot as possible. It is evident that this will not be an even road to travel (Gauquelin himself committed suicide after spending most of his life with scientific research into astrology).

So yes, more scientific research is needed, but that shouldn't stop you from practicing astrology the way you have always done.

Perhaps "proof" was the wrong word to use, Sue. I have a natal Pluto Mars conjunction in Virgo, sextile a Scorpio Moon Neptune conjunction on one side, and a Cancer Mercury on the other side. I suppose, by nature, I need to define things, and that's what I want, rather than proof.

And for all the reasons Peter outlined, I feel astrology needs to have more legitimacy in the scientific community. Access to resources, funds and data are crucial, IMO, to continuing any formal study of astrology. And while I don't believe the scientific approach is necessarily going to ferret out all of astrology's secrets (intuition is an important factor in astrological analysis) I think it can go a long way towards helping us separate demonstrable facets of astrology from the "it works for me" thing.



Thanks again for posting the rest of the interview. A couple of comments:
A: My hope is that, with AstroDatabank, people will be able to start homing in on which are the major factors and which are the fine-tuners, so that we get a more solid sense of what our fundamentals are.
This is exactly what I am doing right now. I am able to see for the first time in an objective manner what the contributing factors are and how strong an effect they have. There are quite a few surprises.
Q: ?If astrology is this skill which people can acquire... then you could cut out the middle-man and just test astrologers...? Would you accept that as a valid line of argument?
A: Yes, this is a valid line of argument.
I have to disagree with this. Putting astrologers to the test is introducing the middle-man. Now you are not testing astrology but the ability of practitioners to apply an astrological body of knowledge.

Other than that, I share Mark McDonough's view of the present state of research and the future with the exception that I am even more optimistic than he is!