The original article in Skeptical Inquirer Winter 1986–87 had a relaxed cover picture by artist Ron Chironna and this introduction by Editor Kendrick Frazier: “We begin publication in this issue of Geoffrey Dean’s two-part ‘Does Astrology Need to Be True?’ a comprehensive investigation of the claims of serious astrology as defined by ‘serious’ astrologers. Although we are striving for shorter articles, so that we can cover a wider range of interests, we publish this lengthy inquiry because of its special significance. As one of our reviewers of Dean’s manuscript wrote, ‘It is without doubt the best article on astrology I have ever seen.’”
Other than its length, the original article had three claims on your rapt attention. It was the result of much recycling among colleagues and noted skeptics, including Susan Blackmore, Ray Hyman, Ivan Kelly, Andrew Neher, and Marcello Truzzi, whose critical comments kept it on the rails. It ignored the nonsense of sun sign astrology and focused on the real thing as used in consulting rooms, on why people believe in it (because it seems to work), and on the results of tests (astrology stops working when cognitive artifacts such as confirmation bias are controlled). And finally it asked, not is astrology true, but does it need to be true? A change that in one hit ended a centuries-old shouting match over claims of truth. The answer was no.
The real thing was not hard to find. In Western countries, it was the subject of roughly 100 periodicals and 1,000 books in print and was practiced or studied by roughly one person in 10,000. (The proportion today has been affected by astrology on the Internet but is probably not much different.) My conclusion in the original article was:
In the last ten years various studies have addressed astrology (the real thing, not popular nonsense) on the astrologer’s terms. The results of these studies are in agreement, and their implications are clear: Astrology does not need to be true in order to work, and contrary to the claims of astrologers authentic birth charts are not essential. What matters is that astrology is believed to be true, and that authentic birth charts are believed to be essential.
Business as Usual
Astrologers replied in their usual way to criticism, dismissing it as biased and ignorant. Their repeated claim—that their daily experience confirms their fundamental premise as above so below—is still heard from the rooftops. They still misinterpret cognitive artifacts in a chart reading as evidence of links with the heavens. And they still explain away all failures by the same old excuses, such as stars incline and do not compel; another factor is interfering (there is always another factor), and astrologers are not infallible. Astrology is thus made nonfalsifiable, whereupon belief and paying clients follow automatically. It then gets worse.
Unwelcome evidence is dismissed because, they say, research is biased; astrology is too nuanced to be testable by science, and (the ultimate clincher) research funding is nonexistent. Yet astrologers insist that looking at birth charts will convince us that astrology works. Just try it and our eyes will be opened at last! But they cannot have it both ways. Astrology cannot simultaneously be difficult to test and yet easy to prove. Their response to this contradiction is usually a scornful silence.
Nevertheless, the past thirty years have seen big advances in research design, the availability of data, and the use of computers to break the calculation barrier. At one time, astrologers using logarithms could take many hours to calculate a comprehensive birth chart; a home computer can do the same for dozens of charts while you cough.
The result has been hundreds of controlled tests of astrology by both believers and critics. Most studies are little-known; so for forty years, my colleagues and I have been visiting astrology collections and searching academic databases for every useful study ever made. We have published the results in Tests of Astrology: A Critical Review of Hundreds of Studies (Dean et al. 2016). Some of the notable studies we found are outlined below. They show how skeptical inquiry has advanced on astrological claims during the last thirty years.
Every astrological consultation involves feedback (as shown above) to help the astrologer pick chart factors that fit the situation. But how accurate are their meanings? The late Dr. Andrew Patterson lectured in engineering at the University of Witwatersrand. His interest in astrology began in the 1960s, and for many years he was a teacher and invigilator in South Africa for the U.K. Faculty of Astrological Studies. His scientific background resulted in that rarest of combinations—a fine critical sense plus an encyclopedic grasp of astrology—which he applies below to the challenge of learning astrology. As you read his account (abridged from 1991), remember he is a teacher of astrology, not a debunking skeptic.
Typical journey from birth chart on left to consultation on right, where the astrologer is famously saying to her client, Adolf Hitler, “With Libra rising you could find great satisfaction in your own home decorating business—or then again you may prefer to invade Poland.”
Astrology is more difficult to learn than anyone realizes. Probably we have all had much the same experience. You meet astrology via a friend and become hooked. You start studying. But after a while you grow uneasy. It is not clear how Sun in Leo (must shine) differs from Moon in Leo (needs to shine). When asked to describe Saturn (restriction) in 8th house (death) you are not sure where to start. All you can say about a hard aspect is that it represents a challenge, whereas an easy aspect is, well, easy. As for a quincunx, you struggle just to pronounce it.
To clear up your confusion, you buy every recommended book. But they just make your confusion worse. Consider the interpretations given in those books. They are either all the same so they blur into one another as with Leo above. Or they are all different, thus Sun square Saturn varies from “a life of hardship” to “loss of father.” Or they are all useless, being either amazingly general or amazingly specific, thus Mars in Libra varies from “lack of committment” to “passion for sword-dancing.” Or they are all evasive as in “Neptune dissolves,” which conveys nothing while pretending to convey everything.
Patterson concludes by pointing out that truth in astrology is tested by how well it matches the symbolism. Anything that passes this test is seen as true, not because it is actually true but because it could be true. Being able to say that the truth (whatever it is) is consistent with the symbolism is not terribly useful. Which is why astrology is so hard to learn (Patterson 1991).
Which Zodiac to Use?
Western astrologers use the tropical zodiac tied to the seasons, while Eastern astrologers use the sidereal zodiac tied to the stars. Around 200 ad, the two zodiacs coincided, but today precession has put sidereal signs almost one sign ahead of tropical signs. So have their meanings changed?
British astrologer J.E. Sunley spent ten years comparing meanings between tropical sign X and sidereal sign X as given in astrology books. In principle, their meanings should be mostly different, but he found they were mostly similar—which is consistent with signs having no meaning at all except in the minds of astrologers. It explains why tens of thousands of Western tropical astrologers can agree that in their experience Scorpio is intense, while hundreds of thousands of Eastern sidereal astrologers can look at much the same piece of sky—which they call Libra—and agree that in their experience it is not intense but relaxed. So much for experience.
But if relative sign meanings are okay, as in Leos get on well with Sagittarians, what is there to worry about?
Sun Signs for Lonely Hearts
Sun sign compatibility was explored by Manchester University’s David Voas (2007) using data gathered for the 2001 census in England and Wales. Traditionally, favorable angles between any two sun signs are said to be the conjunction 0º (Leo and Leo), sextile 60º (Leo and Libra), and the legendary trine 120º (Leo and Sagittarius). Despite possible conflict with other factors in the two charts (among sun sign astrologers this is the default explanation for awkward findings), if the claim is true then it should show up in a large enough sample: ten million marriages, for example.
Voas notes that completion errors are problematic. Census forms are typically completed by one member of the household, who for some reason may enter their own birthday for that of their spouse. Others may enter January 1 or July 1 if an exact birthday is unknown, which is sometimes the case in old people’s homes and for people born overseas. If dates of birth are illegible or missing (about 0.5 percent of all responses), the census office enters the day as the first of the month and assigns the months in rotation. Voas carefully removed all such artifacts but was unable to find evidence for useful sun sign effects.
Astrology predicts compatibility for the angles in red. Most are positive, but none are useful or statistically significant and all are explained by knowledge of sun signs biasing the outcome, as explained in the next section.
Thanks to his enormous sample, Voas’s test was the most sensitive test of sun signs ever made. But none of the 144 possible sun sign pairings differed significantly from chance alone. In terms of predicting compatibility, sun signs absolutely did not work. You will not find this result in astrology books.
Experience 1, Science 0
British astrologer and former journalist Dennis Elwell (1930–2014) was noted for his eloquence. In an article in the Astrological Journal (1991), he restated the faith of astrologers in their experience as follows: “Like many others, I persevere with astrology because experience has shown that by and large its basic assumptions are correct. . . . If some piece of research proves a dead end, I do not question the authenticity of my experience, I question the competence of the research, or its underlying assumptions.”
He held that failures to verify astrological claims were caused by the wrong approach because the right approach always worked. One of his favorite examples was how the birth chart for the Declaration of Independence on July 4, 1776, showed strong links with the Statue of Liberty. Thus the statue is big (Jupiter), made of copper (Venus), has a female form (Venus), and appears in the birth chart as Venus conjunct Jupiter exact within 3º. And so on through dozens of events and associated people. It was “the kind of evidence that astrologers recognise and respect,” and it convinced Elwell (as claimed in his 1987 book Cosmic Loom: The New Science of Astrology, which contains no science despite its title) that “science will eventually be obliged to embrace the astrological if it is to unify its picture of the universe.” So please test astrology by case studies, not by statistical studies of groups.
Okay, let’s do it. Suppose we’ve been told that the above chart has its Sun conjunct Uranus, whose meaning “very frequently indicative of great talent” could hardly be more apt—is astrology already discernible? Indeed, the statue is an innovative (Uranus) national monument resting on Sun-ruled granite and lit by electricity (Uranus). It is 151 feet high (equals 1º Leo, which sign is ruled by the Sun) on an eleven-pointed island (obviously the eleventh sign Aquarius, ruled by Uranus). Everywhere we look we find the predicted Sun-Uranus links. Yes, it’s amazing!
Since this is “the kind of evidence that astrologers recognise and respect,” we now have good reason to believe in astrology—except the chart has no actual Sun-Uranus conjunction (Uranus is 40º from the Sun, not 0º or any other aspect within traditional limits). Elwell’s respected evidence is no evidence at all.
The correlation between sun sign and extroversion scores is usually weakly positive and was once hailed as proof of astrology. But it disappears (red crosses) if subjects don’t know sun sign meanings or are tested against their moon sign (which few people are aware of). So the effect is in fact an artifact of sun sign knowledge.
Sun Signs and Self-image
Odd-numbered signs from Aries onward are said to be extroverted. The rest are said to be introverted. Ask Sagittarians (odd-numbered and said to be sociable and outgoing) a question related to extroversion (such as “Do you like parties?”), and knowledge of astrology might tip their answer in favor of yes rather than no. In fact, this answer-tipping can be detected if people know their sun sign but not if they don’t. When taken together with opinion polls, the results suggest that one in three people believes sufficiently in sun signs to measurably shift their self-image in the believed direction—of which a tiny fraction may believe sufficiently to bias their choice of partner as in the previous section.
Astrologers Put to the Test
Charles Carter, the leading British astrologer of the 1930s, was noted for exceptional clarity of expression. Here is an example from his book The Principles of Astrology (1925, 14): “Practical experiment will soon convince the most sceptical that the bodies of the solar system indicate, if they do not actually produce, changes in: (1) Our minds. (2) Our feelings and emotions. (3) Our physical bodies. (4) Our external … affairs and relationships with the world at large.”
Thirty years ago, such claims began to be tested by jumbling up birth charts with things such as their owner’s case histories and personality traits. Could astrologers match them correctly? The outcome was maybe yes but mostly no. Since then, more tests have been made that bring the total to sixty-nine, and new ways have been developed to analyze the results. For example, the correlation between a reading and reality can be plotted against sample size to clarify what is happening. The plots in Figure 1 show how it works.
Figure 1. Samples from uncorrelated data (r = 0.00) have sampling errors that produce non- zero correlations (r ≠ 0.00) especially for small samples (common in astrology). Right: If astrol- ogers could accurately match birth charts to their owners then the black dots would peak on the right. But they peak close to r = 0.0, or zero accuracy, and are skewed to the right indicating the presence of publication bias against nega- tive results, hence the slightly positive mean (r = 0.029). Here and in Figure 2, all means are weighted by sample size. Discarding tests with small sample sizes or less familiar criteria makes no difference.
The studies in Figure 1 are too numerous and too consistent with hundreds of other studies to be easily dismissed. Also, their subsequent meta-analysis shows that the differences between results are entirely explained by sampling errors, which leaves nothing for astrology and astrologers to explain; to paraphrase Pierre-Simon Laplace, we have no need of such hypotheses. But for completeness, we should still look at some of those other studies as shown in Figure 2.
Figure 2. Tests of agreement when reading the same chart avoid all problems of judging accuracy, such as errors in birth time. The agree- ment should be high because astrologers tend to read the same books, but it is only weakly positive (r = 0.098) and is nowhere near the 0.8 generally required for tests applied to individu- als (as astrology is). What is astrology worth if astrologers cannot even agree on what a chart means? Right: Clients given several chart read- ings cannot pick their own reading unless it contains cues (as required by the experimental design), such as the name of their sun sign, in which case they do quite well.
The power and sensitivity of our tests so far are beyond anything the ancients could have dreamed of. But astrologers airily dismiss the results because, as one put it, “We have enough cumulative experience to know that it [astrology] works, whether the computer studies and the scientists agree with us or not” (Alexander 1983, xii).
Claims Tested on 3,290 People
For his PhD in psychology, German astrologer and psychotherapist Peter Niehenke (1984) circulated copies of a 425-item questionnaire for testing astrological claims. It was advertised in two newspapers and a New Age magazine and by notices at Freiburg University. He duly received 3,498 responses (requiring more than 110 reams of paper), of which 3,290 provided usable birth data, of which 62 percent were from birth certificates. The questions had been tested in a pilot study to make sure they were free of problems. Each was relevant to a given factor (planet, sign, house, or aspect) to see if the subjects identified with that factor regardless of whether it was actually in their birth chart. Thus Sun-Saturn aspects were explored by questions involving their supposed meanings such as disappointment, misfortune, pessimism, and guilt feelings.
Overall, no result was consistently in support of astrology. For example, subjects with four Saturn aspects (said to indicate heavy responsibility and depression) felt no more depressed than those with no Saturn aspects and showed no correlation with depression scores. Subjects with good trines to Jupiter (said to indicate optimism and good fortune) felt no sunnier than those with none. Aspects between aggressive Mars and the Sun, Moon, or Ascendant showed no correlation with aggressiveness scores. Responses to the question “I am unlucky in love: yes/no?” showed no correlation with aspects to Venus from Jupiter or Saturn, or with the house position of Saturn, all of which are said to be highly relevant. In the end, Niehenke decided there was more to astrology than being true or false: “a world in which astrology exists is surely a more enjoyable world than one without it. The need that astrology be a reality is much stronger than all the rational demonstrations against it” (1984, 15).
300,000 Chart Factors
In 1996, U.S. database engineer Mark McDonough wrote software to store and deliver the 30,000 birth data in AstroDataBank, the world’s second largest collection of timed birth data. After several years of work, he could automatically analyze any subset of data for 300,000 chart factors (that’s not a misprint; the large number is due to fashionable ideas such as asteroids and planetary nodes) taken individually or in combination and identify which factors differed the most from controls. But when applied to actual birth data grouped by, say, occupation or events, the results if positive (which was not often) failed to replicate. There was no evidence that astrological claims were valid: nothing actually worked. He asked for an explanation, but nobody had a clue. So he abandoned astrology to follow other interests.
Wrong Charts Make No Difference
Do astrologers get right answers from wrong charts? If they do, then their fundemental premise as above so below is disconfirmed. The idea might seem difficult to test—what astrologer wants to read wrong charts?—but it happens purely by accident and is surprisingly common. The astrologer gives a reading that satisfies the client but the wrong chart has been used. It makes no difference how wrong it is—by hours, days, or years—the chart still works. Astrologers recognize this but see it as some occult property of astrology that puts it beyond human understanding. Skeptics may disagree.
The Gauquelins in 1981
Les Gauquelins et leur Héroïsme
The most heroic studies in astrology were made by French psychologists Michel Gauquelin (1928–1991) and his wife Françoise (1929–2007). They used statistical testing and large samples mostly from the nineteenth century. Their results for traditional astrology (signs, aspects, transits) were consistently negative. Nothing worked. Therefore they were surprised to obtain positive results for what was later called the Mars effect (and, later still, planetary effects because the Moon, Venus, Jupiter, and Saturn were also involved): the tendency for eminent professionals to be born when the planet matching their occupation (such as Mars for sports champions, Jupiter for actors) had just risen or culminated. Planetary effects were new in that, unlike previous factors, they were critically dependent on the hour of birth.
Statistically, the effects were often very significant, which to astrologers meant strength. But their effect sizes, which for over thirty years nobody bothered to calculate, averaged a tiny r = 0.04 ignoring direction. So the effects were actually weak and were significant only because large samples were tested (typically more than 1,000). Indeed, the effects were so weak that if applied to 100 clients, on average only two would get readings more accurate than tossing a coin—and even then only if they were among the one in 20,000 who were eminent. Yet the effects replicated and were not explainable by faulty procedures (see Figure 3).
Figure 3. Effect sizes for the fifty-nine known studies based on computer-calculated data spread over five planets—Moon 12 percent, Venus 4 percent, Mars 38 percent, Jupiter 25 percent, Saturn 21 percent. Some results have negative effect sizes (e.g., Saturn for painters), so they do not cancel out positive effect sizes. What matters is the proportion of effect sizes that lie outside the 50 percent confidence limits. If significantly more than 50 percent, as is the case here, then planetary effects seem to be real.
Ironically, planetary effects created baffling puzzles even for astrology. Why only five planets? Why no effect for the sun or for signs and aspects? Why occupation and not personality? Why contrary to all expectations are planetary effects larger for less-precise birth times? And why are there such strange effects in the first place?
For forty years, nobody had a clue. Astrologers predictably saw the effect as proof of the higher realities in which astrology is said to operate. But after eight years of work, I uncovered a new artifact capable of explaining all the puzzles—namely the misreporting of birth times to match the pop astrology of the day (Dean 2002). The level of misreporting was very small, but then again so were the planetary effect sizes—and as opportunities for misreporting disappeared, so did planetary effects. Nobody knows if planetary effects still apply today, but that’s only because privacy laws make new data hard to find. In any case, planetary effects are far too weak to be of practical use to astrologers.
But might consolation be found in Indian astrology, claimed by Indian astrologers to be vastly better than anything available in the West?
Last of the Astro Mohicans: Gauquelin planetary effects were the last of the astrological areas that had seemed promising in the 1970s and which still remained when my original SI article was published in 1986. The indicated end year is when the prom- ise was lost due to the discovery of artifacts—poor control of chemistry (Kolisko), biased samples (Jung), inappropriate tests (Nelson), social arti- facts (Gauquelin), astronomical artifacts (Bradley), sampling errors (Addey and Clark), unavailable data (Jonas), and sun sign knowledge (Mayo-Eysenck).
An Indian Test of Indian Astrology
Indian astrology is hugely different from Western astrology. It is more complex, uses the sidereal zodiac, and fortune-telling is the norm. The scientific revolution that eroded astrology in seventeenth century Europe did not happen in India, so it has had a free run ever since. Today it is firmly entrenched at all levels of Indian society. But no controlled test had been made in India until the one by Jayant Narlikar and colleagues at the Inter-University Centre for Astronomy and Astrophysics in Pune (Narlikar 2013).
They gave each of twenty-seven volunteering Indian astrologers (mean experience fourteen years) a different set of forty timed charts each, and a team of astrologers 200 timed charts (a larger number than in any Western test), to see if they could tell bright children from mentally retarded children. This is a commonly accepted claim in India, but neither group outperformed tossing a coin.
Nightmare on Time-Twin Street
“Time twins” are people born close enough in time and geography to have similar birth charts. At a given moment, the birth chart supposedly indicates trait X, the next moment it is trait Y, and so on. So time twins should be more alike in X than expected by chance, which makes them the definitive test of astrology, since all confounding reading artifacts are avoided.
In a city of one million people, more than 2 percent will have a time twin born within one minute, about the same proportion as people with an ordinary twin, and about 20 percent will have a time twin born within ten minutes. The numbers increase very rapidly with time difference and city size. Indeed, the number of time twins in Western history is so enormous (hundreds of millions) that many similarities in personality and events will occur by chance alone. So the handful of cases routinely cited by astrologers cannot hope to be convincing.
Distribution of 5,591 NCDS births in Southeast England during March 3–9, 1958 with the expected peak in the Greater London area.
The systematic testing of time twins was explored by Ivan Kelly and me (2003) using cohorts from the National Child Development Study (NCDS) of 16,000 children born in the United Kingdom during March 3–9, 1958. To minimize variations in birth place, we analyzed only those born in Greater London. Birth times for 92 percent of cases were reported to the nearest five minutes, and the rest to the nearest minute. For each person, we selected a total of 110 variables measured at ages eleven, sixteen, and twenty-three that were said to be shown in the birth chart such as ability, accident proneness, behavior, occupation, personality, and physical data such as height, weight, vision, and hearing. Data collection had required whole armies of researchers well beyond anything astrologers could achieve. For the purposes of testing astrology, this database was a dream come true.
But for astrology itself the results were a nightmare: support for astrological claims was nowhere in sight. For example, Saturn sets every day momentarily exactly on the horizon, a position traditionally held to greatly boost its strength. At that time in London on March 6, 1958, it was also square the Moon within 0.1º, which is also held to boost its strength. It was not just a strong Saturn event; it was also the strongest Saturn event for the entire week. Saturn is held to indicate restriction and limitation, so its effect should show up as a dip in measures of ability. But it did not (see Figure 4).
Figure 4.
A Strong Saturn Fails to
Show Saturn Effects
Ability scores (a composite of fifteen tests such as intelligence, reading, and mathematics) plotted against time of birth for 2,193 births (Figure 4) shows no discernible effect from the Saturn event, no daily rhythm that might coincide with rising or culminating planets as in Gauquelin’s planetary effects, and no clear difference from the same data when the birth times are randomized (lower plot). The white lines are forty-one-point moving averages (forty-one points is about three hours). None of the other 110 variables fared any better when analyzed by a battery of tests. But can we be sure that the test is really appropriate? It may be that ability is too broad a measure to show Saturn effects, in which case we need something such as extroversion that is more definitely linked to Saturn (caution, reserve). Perhaps Saturn effects are too focused to be discernible during seven days, in which case we need a smaller time frame. All as in the next test (see Figure 5).
Figure 5.
No Saturn Effect on Extroversion
Astrology predicts a drop in extroversion scores (here based on ratings on thirteen relevant scales such as impulsive–cautious) below the mean during the Saturn event (black dots in Figure 5). But if anything they increase, albeit not significantly (p by a t-test is 0.22). The extroversion scores show no tendency to group together. Enlarging the Saturn event window makes no difference, so the time twin similarities predicted by astrology are not detectable.
Could Lack of Resources Be the Problem?
At this point, the last hope for astrology’s factual validity seems to disappear, but there are still straws to clutch at. Astrologers claim that with enough funding, research facilities, and right-minded researchers, astrology would soon regain its rightful place as queen of modern worldviews. This belief has been put to the test in each of more than a dozen PhD theses that have involved tests of astrology.
It did not work for Niehenke’s PhD thesis. So let’s look at the PhD thesis of Pat Harris, a British astrologer whose website offers you a £30 ($45) Astro Fashion Profile based on sun signs. Earlier, during twenty years of professional practice, one of her clients had conceived via IVF (in vitro fertilization) under astrological conditions that were absent during seven failed attempts. This seemed to suggest that astrology could improve the IVF success rate—an idea she explores in her thesis (Harris 2005).
Later in an article in the Journal of Sexuality, Reproduction & Menopause (2008), Harris claimed that “attempts to conceive during [astrologically] optimal times have an increased likelihood of success,” even though an editorial note advised that her results were not statistically significant. The May 2009 issue contained a letter from Jacky Boivin, professor of health psychology at Cardiff University, who noted that Harris’s two samples of twenty-seven and twenty-eight women were too small to escape sampling artifacts (for which about 400 would be needed), thus her claim “is completely unwarranted.”
Unusually, her thesis was under an embargo (normally granted only if it contains commercially sensitive material) that for five years prevented its release. In due course, I found it to contain no birth data, no proper controls, no expectancies, no details that would allow an independent check, and success rates inflated in much the same way as predicting dryness in arid areas—exactly the sort of errors and omissions that in my day would get your thesis rejected. What was her university thinking?
Like Niehenke, Harris did not let her results influence her belief in astrology. And here we encounter astrology’s dark side—on another website she now offers for £90 ($135) the best astrological dates for achieving conception plus a £148 ($220) telephone analysis of your birth chart to optimize fertility. In her 2009 letter, Professor Boivin had commented that Harris’s paper “should not have been published because it falls short of the scientific standards adopted to create the evidence base for interventions in fertility. . . . People with fertility problems are willing to try anything to achieve pregnancy, and giving them false hopes is yet another way of taking advantage of this vulnerability.”
This of course calls into serious question the scientific and ethical standards of Harris’s actions. So let’s try one more time with the PhD thesis of Keith Burke (2012), a former U.S. astrologer who went further than most by cofounding a for-profit institute for teaching New Age topics. He taught astrology classes and held workshops through the institute, wrote astrology articles and a textbook, and lectured at national conferences. Verily the definitive right-minded researcher!
He had noted that the Moon is generally held to be as important as the Sun but had received little attention by researchers. There was also a clear similarity between the Moon’s meaning in each of the four astrological elements and four of the Big Five personality dimensions. So this became the subject of his PhD thesis at the Pacifica Graduate Institute, an accredited clinical training graduate school in California that was even better suited to astrology than a pure research school. According to astrologers, the results ought to support astrology. But the effect sizes for 192 subjects with timed births, mean age forty-nine, were not only at chance level, but three were in the wrong direction (see table below).
Element
Big Five
r
p
Fire
Extroversion
- 0.082
0.49
Earth
Conscientious
- 0.006
0.27
Air
Intellect
- 0.074
0.31
Water
Neuroticism
0.050
0.94
The funding, research facilities, and right-mindedness (to say nothing of a promising hypothesis) had been to no avail. Unlike Niehenke and Harris, Burke had already stopped reading charts for clients, a decision helped by his concerns about people looking not for counseling but for major life answers that a chart cannot give. He is now a clinical psychologist and a professor of behavioral sciences, and he does not use astrology in his profession or personal life (Burke 2015).
The case for and against astrology can now be briefly stated. Since thirty years ago, the case against has become stronger. The case for remains unchanged.
Cases for and against Astrology
Astrology is among the most enduring of human beliefs and has undisputed historical importance. A warm and sympathetic astrologer can provide wisdom and therapy by conversation with great commitment that in today’s society can be hard to find. To many people, astrology is a wonderful thing: a complex and beautiful construct that draws their attention to the heavens, making them feel they are an important part of the universe. However, to their discredit, astrologers fail to recognize astrology’s many problems. They refuse to accept that experience can be unreliable; they brush aside negative evidence; and they dismiss critics as close-minded by definition. As a result, astrologers are promoting both an illusion and a deceit. They are astrology’s own worst enemies. Ultimately, the issue is a personal one—whether factual truth is to be more important than personal meaning. Skeptics will no doubt have thoughtful responses to that one. Answer to the Title QuestionThe tests outlined here lead to the same answer as do hundreds of other tests. They confirm that nothing in a birth chart is sufficiently true to support the meanings claimed by astrologers. Their books, classes, and conferences are not built on evidence but on opinions based on opinions based on opinions, thus perpetuating the seeing of faces in clouds. Millennia have not wearied them.
So the answer to the thirty-year-old question in the title remains the same. No, astrology does not need to be true in order to seem to work. It is simply a time-honored cover for artifacts that better explain the outcomes. Astrologers have had ample opportunity to prove otherwise by controlled tests but have not done so, a failure most easily explained by their being unable to do so. As a consequence, astrologers should not be surprised if they find themselves disqualified from positions of credibility in Western society.
Nevertheless, depending on who we are, we can still see astrology as beautiful, spiritual, helpful, controlling, lucrative, great fun, or simply stupid. But one final question.
How to React?
French social scientist Laurent Puech (2003, 267), in a book-length study of the pretensions of astrology, suggests that the best reaction to astrology lies in the provision of reliable information and critical tools:Whether we like it or not, astrology and recourse to astrologers is here to stay. I think they will never disappear because they fill a need. They will be simply more or less important according to the times. How to react? . . . [It is] not a question of censuring astrology but of helping people to find reliable information about it, and also to find the minimum critical tools for evaluating it.
The problem for astrologers who wish to promote their invalid views of astrology is how to stop people from finding out the truth, even though some may see astrology as having more to it than being true or false. n
Acknowledgements
My thanks to Susan Blackmore, Wout Heukelom, Ivan Kelly, Arthur Mather, David Nias, and Rudolf Smit for helpful comments—for most of them a heroic repeat of their help with the original article of thirty years ago.
References
Alexander, R. 1983. The Astrology of Choice. New York, NY: Weiser.
Burke, K. 2012. Big Five Personality Traits and Astrology: The Relationship between the Moon Variable and the NEO PI-R [Big Five personality inventory]. PhD thesis, Pacifica Graduate Institute, California.
———. 2015. Personal communication with the author.
Carter, C. 1925. The Principles of Astrology. London: Theosophical Publishing House.
Dean, G. 2002. Is the Mars effect a social effect? Skeptical Inquirer 26(3) (May/June): 33–38.
Dean, G., and I.W. Kelly. 2003. Is astrology relevant to consciousness and psi? Journal of Consciousness Studies 10(6, 7): 175–198.
Dean, G., A. Mather, D. Nias, et al. 2016. Tests of Astrology: A Critical Review of Hundreds of Studies. Amsterdam: AinO Publications. For details, visit http://www.astrology-and-science.com.
Elwell, D. 1987. Cosmic Loom: The New Science of Astrology. London: Unwin Hyman
.
———. 1991. Can astrology win through? Astrological Journal 33(1): 49–50.
Harris, P. 2005. Applications of Astrology to Health Psychology. PhD thesis, School of Social Sciences, University of Southampton.
———. 2008. Managing fertility treatments and stress with astrology. Journal of Reproduction, Sexuality, and Menopause 6(3): 43–44.
Narlikar, J.V. 2013. An Indian test of Indian astrology. Skeptical Inquirer 37(2) (March/April): 45–49.
Niehenke, P. 1984. The validity of astrological aspects: An empirical inquiry. Astro-Psychological Problems 2(3): 10–15.
———.1987. Kritische Astrologie, 194. Freiburg: Aurum.
Patterson, A. 1991. Why astrology is so hard to learn. Considerations 6(3), 5–13.
Puech, L. 2003. Astrologie: Derrière les mots [Astrology: Behind the Words]. Sophia Antipolis: Éditions book-e-book.
Voas, D. 2007. Ten Million Marriages: A Test of Astrological Love Signs. https://astrologia
experimental.files.wordpress.com/2014/05/voas astrology.pdf. A version without tables is available in Skeptical Inquirer 32(2): 52–55, Ten million marriages: An astrological detective story.