Tuesday, January 5, 2016

In defense of astrology

Astrology is the study of personality and personal events with respect to celestial objects and ones birthdate and location. It is often used in prediction of events of a person's life and their personality. In the European tradition, astrology is mostly based on the position of the sun in the sky, which constellation of the zodiac it is in, on a person's birthdate.

Most people are aware of astrology through the idea of a daily horoscope (often printed in newspapers on the comics page next to the word jumble or chess column) (these days with the increasing marginalization of newspapers, I don't know where people experience it).

Many people view it, and dismiss it, as entertainment, on par with palm reading or Tarot cards, not something to be taken literally. Those with a scientific or practical bent might dismiss it outright as an empty myth or BS because it is, to use the jargon, not falsifiable. There are a some who still believe that it may work.

But the general consensus is that it is BS. All I want to do is pedantically (but not to contradict this consensus which I share) show that Astrology is indeed scientific in the technical sense, just one that is not supported by the data. A science that has been falsified is still scientific, just disproven.

There are two common applications of astrology. One, based on your birth sign one out of twelve, to predict, is very vaguely the events of the day. For example (taken from a random horoscope site; it is very exemplary of the kinds of writing):
Today you urgently need to understand and master the art of balancing the physical reality with your vision. While your plans are ambitious, you have to understand the actual obstacles top these plans. Otherwise, you are headed for a collision course in spite of all your good intentions. You also need to understand that your plans may be conflicting with those of someone else who is as determined and ambitious as you.
(source (no link to the specific text for a day, just generic for the sign))

These tend to be the greatest source of disbelief: how can everybody having the same birth sign have all those things happen? These read like fortune cookies that could be true of anyone whatever their birth sign. Accidentally read the fortune of another sign and that is just as likely to happen or not. Astrological predictions like the above are a great example of confirmation bias in action; all you need to happen is for one thing to match what happened that day (if you read the horoscope after the day), or you just may make one of things happen, and you'll come to 'believe' that horoscopes work.

The other common application is a prediction of your distinct personality. A natal chart, the position of major celestial bodies with respect to the constellations and each other at the the time and place of birth, is used to predict personality. The position of the sun is the most important, telling you your 'sign', but the moon and planets and their relations are important. This facts lead to statements of the form like 'The moon is on the cusp of Libra and Virgo, and Mars is in opposition to Venus but square with Saturn meaning that...' followed by some explanation involving the personalities of those celestial bodies, inspired by their mythological stories.


(image from wikipedia)

My defense is that all these things make astrology not simply a myth but an actual scientific theory, scientific in the sense that it could be the case, and that we can test whether predictions by it or its parts hold or not and there is a postulated mechanism for these effects that holds up. Of course, none of it holds at all. My defense is simply that astrology is legitimately falsifiable.

Counter to most opinions, I consider astrology to be falsifiable because one can make a study of daily horoscopes or a study of star charts and individual's life events or personalities. There is very scientific measurement of celestial body positions. There can be predictive personality tests (but many currently personality tests are currently not considered particularly good at prediction, for example Meyers-Briggs personality test (MBTI), the somewhat similar Big Five test is supposedly somewhat more predictive).

St. Augustine's dismissal of astrology because twins don't have identical outcomes is motivating to me but scientifically could be problematic. Twins do share quite bit of personality. This thought experiment this could be modified only the slightest to be more supportable; compare the lives of two children born the same day in the same hospital. The general predictions of 'astrological theory' can barely be supported. Some lifestyle and class similarities will predict most of the similarity between the two children.

The horoscopes in newspapers or your astrological sign ("You must be a Libra!") are, I agree, not scientific because they are so vague as to be barely even functional much less predictive (meaning that the pronouncements they make can barely be mapped coherently to life events or personality.

But an astrology based on actual measurement could be falsified and therefore is a scientific theory. Its effects just have never been established.

By the way, the Middle-eastern-based astronomical astrology (the greco-roman constellations-sun-moon-planets and mythically associated personalities) may have exact predictive power but there's one huge glaring plain old mistake in execution. The constellation the sun is in, one's sun-sign, as usually given today, is mismatched with the prescribed dates.  The twelve ranges of dates are set on the sun based along the ecliptic and the positions of the constellations. The constellation labels were set 2000 years ago. Because of the (astronomically measured) precession of the equinoxes, the position of the sun during the year has shifted a little more than the space of one constellation.



(source: wikipedia)
That is, on March 21st (the spring equinox, that start of the zodiacal year), when tradition states that the sign of Aries begins, in the year 2000 the sun is actually as measured in the sky against the backdrop of the constellations, already part way through the next constellation of Taurus.

So this is not an error of interpretation or logic, it is an even more elementary error of data recording/reporting.

No comments: