Thursday, June 17, 2010

Is Astrology Scientifically Verifiable? A Review of “Astrology: Science or Superstition” by Eysenck and Nias

Hans Eysenck and David Nias take on the contentious issue of whether or not astrological ideas can be scientifically verified in their book Astrology: Science or Superstition. This book, published in 1982, goes over a lot of research, both of traditional astrology and of a newer quasi-astrology called “cosmobiology.” While the book is almost 30 years old, I decided to read it because I admire and enjoy reading the works of Hans Eysenck.

Hans Eysenck (1916-1997) is my favorite 20th century psychologist. While not as well known or influential as Freud or Skinner, I think that Eysenck was on the right track in his research and theories. During the mid-twentieth century, when unproved psychoanalytical theories were predominant, Eysenck was a fervent critic. Eysenck did groundbreaking work in personality theory, backed by extensive empirical research. He’s most famous for his theory breaking personality down into three fundamental categories: extraversion, introversion, and psychoticism. He was a pioneer in the study of the biological basis of personality, intelligence, and mental health. Not one to shy from controversy, his view that racial differences in IQ are (partially) genetically determined got him punched in the face during a talk.

Most scientists avoid studying astrology, considering research in the field to be a waste of time and resources, kind of like studying psychic phenomena or human magnetoreception. But what if there is something to it, something that we don’t understand very well at present, but may lead to a new understanding of human behavior? It must be remembered that not long ago, the idea that some psychiatric disorders have a genetic component would have been laughed off as ridiculous. The same with the idea that taking pills would treat symptoms of depression or schizophrenia. What about splitting the atom, landing a man on the moon, or video conferencing with someone halfway around the world? Things that are common knowledge or obvious now weren’t always so.

Eysenck and Nias begin their book by discussing how mainstream scientists are irrationally hostile to anything connected to astrology. They then describe astrology, and delve into the research on the subject. They don’t find much research support for traditional astrology, e.g. whether birth charts and sun signs have any connection to personality, occupation, or personal destiny. Any studies that seemed to show a positive result either had methodological or statistical problems, or failed in replication. While many astrologers are sincere and mean well, any advice they give people isn’t based on research or science.

The authors then turn to what they call “cosmobiology”, or the scientific study of how extra-terrestrial factors influence living organisms. Much of the research in this field consists of initial studies that have not been replicated, and thus can’t be considered to be proven. I’ll omit these studies and focus on the few findings that have been replicated and that Eysenck and Nias consider to be solid. One is that eminent people are born more often between the winter solstice and spring equinox. (“Eminent people” are those who have accomplished enough to be listed in one or more encyclopedias). One study showed 36 eminent people born per day during the peak in February, compared to 27 people per day at the trough in June. There’s a similar trend for schizophrenia. A 1998 study confirmed this excess of winter births for schizophrenia, and also for bipolar and unipolar depression. Eysenck’s book on creativity explains that psychosis and creativity are biologically linked, and the seasonal of birth data support his hypothesis.

One possible explanation for the season of birth data is that the parents of eminent or mentally ill children have a greater tendency for conceiving a child in the spring than the average parent. Why this is the case isn’t clear. Another explanation for the greater prevalence of mentally ill winter births is that babies born during the winter are more susceptible to infections, and these infections cause their mental illness. There’s been no definitive link, however, between any known kind of infection and schizophrenia or mood disorder. This viral hypothesis doesn’t explain why eminent people should be born more often in the winter.

Eysenck and Nias take up the question of whether the sunspot cycle correlates with biological or historical cycles, but don’t find any solid evidence in favor of this. In his later book on creativity, however, Eysenck mentions Ertel’s research. Ertel found a correlation between the sunspot cycle and creative achievement in the arts and sciences. There’s more worldwide creative achievement in the nadir of the solar cycle as compared to the peak. There were also worldwide bursts of creativity during periods in which there were few or no sunspots, such as the Maunder Minimum (1620 – 1710). To my knowledge, Ertel’s research has not been replicated, so this isn’t solid evidence by Eysenck’s and Nias’ standards. Eysenck, however, thought it significant enough in his creativity book to devote 7 pages in explaining it.

The most solid evidence Eysenck and Nias come up with, to which they devote an entire chapter, is the research done by Michel Gauquelin (1928 – 1991), i.e. the Mars Effect. Gauquelin researched the connection between planetary position at time of birth and eminence. From the perspective of the rotating earth, planets rise and set, as do the sun and the moon. Gauquelin divided each planets’ path into 12 sectors, similar to, but not identical with, astrological houses. Gauquelin studied birth records of eminent doctors, artists, scientists, and athletes. He found that eminent scientists tend to be born more often just after the rise (i.e. when the planet first appears on the horizon) or upper culmination (i.e. the highest point) of Saturn, while eminent artists are less likely to be born during these times. He found that eminent doctors are more likely to be born with Mars or Saturn in the sectors following the rise or upper culmination. He found that for eminent military leaders and for “iron willed” athletes, Mars was more likely to be in the sector following the rise or upper culmination (remember that Mars is the ancient god of war).

Eysenck and Nias consider the Mars Effect to be successfully replicated. Since the book is 30 years old, I checked out more recent sources. A 2005 article describes the history of the controversy. To summarize, the jury is still out whether or not this effect is a statistical artifact or not. The previously mentioned Ertel became involved in the controversy, claiming that the Mars Effect did exist in the replications, if one focuses only on eminent people. For example, only highly successful athletes show this effect, not average ones.

Notice how eminence is an important factor in the Mars Effect, as it is in Ertel’s sunspot cycle research, and as it is in the season of birth research. Perhaps eminent people have some type of capability that others don’t have. This capability may be related to the magnetic sense, as the geomagnetic field is known to be influenced by sunspots and other extraterrestrial factors.

I can see a similarity between the controversy about the Mars Effect, and the controversy about Robin Baker’s human magnetoreception research. In both cases, one side argues that the replications were successful, and the other side argues that they weren’t. The key questions are methodological and statistical in nature. Mainstream science is not convinced, and the research project becomes abandoned.

My own opinion about the Mars Effect, echoing that of many scientists, is “so what?” Even if the Mars Effect exists, it can’t be used to predict anything. It’s an interesting correlation, showing that birth times may be connected to extraterrestrial events, but doesn’t really tell much about human psychology or human behavior. Eysenck would probably argue that it’s difficult in general to make predictions in the social sciences. There are so many influences and factors involved in human behavior, that correlational research is often the only method that can be used. The size of the correlations in the Mars Effect is as high as in other mainstream social science research. So why dismiss the Mars Effect, solely on the grounds that we don’t understand how it can work?

Eysenck and Nias make the analogy to Newton’s Law of Gravity. Newton didn’t understand how gravity worked, but was able to construct a successful theory of gravitation. We don’t understand how cosmobiology works, but why can’t we take it seriously, do the necessary research, and come up with a theory? In their concluding chapter, the authors make the connection between cosmobiology and geomagnetism. It’s likely that any extraterrestrial effects are mediated by the geomagnetic field. Research in animal and human magnetoreception is directly relevant to cosmobiology.

I would like to see more research in cosmobiology, but would like to see us move away from the correlational in favor of the experimental. Robin Baker did experimental research on human magnetoreception, but the effect sizes were so small and difficult to reproduce that he was unable to convince other scientists. My human magnetoreception hypothesis points the way to another research paradigm that may have more successful results. I (and some others) have a limited functionality GPS. While this GPS doesn’t provide a great deal of navigational information, the limited information it does provide can be used as the basis for experimental tests. I have the ability to distinguish between being north or south of magnetic home. This ability can be tested in a double-blind manner, by driving me around in a bus with a sunroof and covered or blocked windows. If my sleeping behavior is experimentally controlled as I describe in my research paper, I will be near-perfect in my ability to distinguish between being north or south of home. The effect won’t be masked by statistical noise, as it was in Baker’s experiments and in the various correlational research studies mentioned in this book.

In conclusion, I recommend this book as an accessible, well-written discussion of astrological research.