The case for astrology

An expanded abstract of the article in Journal of Consciousness Studies Volume 10 (6-7), June-July 2003, pages 175-198 with four tables and 85 references. This particular issue of JCS is devoted to parapsychology and related matters, and is also available in book form. Details, abstracts, and the full article in pdf format are available at www.imprint.co.uk/books/psi.html. See also the user-friendly www.astrology-and-science.com/ for critical articles on astrology by the same authors and others.

Why astrology?

Astrology has one sure thing in common with parapsychology – a highly visible outpouring of market-driven nonsense that threatens to bury the work of serious researchers. Just as parapsychology to the ordinary person means ghost busting and psychic phonelines, so astrology means sun signs and newspaper columns. Here we ignore the latter view in favour of serious astrology, the study of purported relationships between the heavens and earthly affairs.

Astrology can be applied to anything that is born or begins independent existence such as a person, company, ship, nation, animal, or idea, and the astrologer begins by calculating the birth chart or horoscope, a stylised map of the heavens at the moment of birth as seen from the place of birth (think of a wheel covered in strange symbols). Then comes the difficult part, namely the interpretation or birth chart reading, where the strange symbols and their positions are translated into a discourse that the client can understand. It is here that the possible relevance of astrology to psi becomes apparent.

Many astrologers attribute a successful birth chart reading to what they variously call intuition or psychic ability, where the birth chart acts like a crystal ball. As in shamanism, they relate consciousness to a transcendent metaphysical reality that, if true, might require a re-assessment of present biological theories of consciousness.

Cynics might argue that the prospects are not good. Even after centuries of practice, astrologers still cannot agree on what a birth chart should contain, how it should be interpreted, or what it should reveal. Nor do they agree on how astrology should be tested, or even on whether it can be tested in the first place. They are also generally unaware of the many hidden persuaders such as the Barnum effect (reading specifics into generalities) and nonfalsifiability (nothing can count against your idea) that can make astrologers see hits where none exist. Technically these hidden persuaders can be described as “statistical artifacts and inferential biases”.

Indeed, it is not uncommon for astrologers to make a seemingly accurate reading only to discover it was based on the wrong birth chart, which they then attribute to some mysterious property of astrology. It seems that critical thinking, and being informed about relevant research, is not for them.

Nevertheless let us accept that astrologers may use some sort of intuition or psychic ability when reading a birth chart. Because the incidence of astrologers and serious students of astrology is roughly 1 in 10,000 of the general population, their total number is substantial and our chance of detecting psi and anomalous states of consciousness is correspondingly increased, which is an opportunity not to be lightly passed by. On the other hand, their 0.01% incidence is very much less than the 4% incidence of fantasy-prone personality (one that fantasises vividly during much of waking life), so who knows?

Empirical studies of astrology

Although very few empirical studies of astrology existed before 1950, their number now exceeds five hundred. Their results have revolutionised our understanding of astrology, and coherent reviews are now possible. In what follows we look first at the performance of astrology, for which the definitive test is time twins, and then at the performance of astrologers. If astrologers can do better than astrology would seem to allow then we might be on to something.

Suppose that at one moment the heavens signify that people born at that moment will have characteristic A, the next moment it is characteristic B, and so on. Time twins (people born close together in time) should therefore be more alike than expected by chance, and more alike than people born further apart in time. Time twins are thus the definitive test of astrology because errors or uncertainties of birth chart interpretation are avoided.

The first systematic study of time twins was reported by British astrologers Peter Roberts and Helen Greengrass in 1994. Their sample of 128 people showed some evidence of similarities in interests and occupation, for example two born 15 minutes apart were respectively a bassoon player and a clarinet player, but there were no clear similarities in appearance, handwriting, names, or life events. The strong similarities predicted by astrology were simply not there.

We conducted a more powerful test involving 2101 persons born in London during 3-9 March 1958 averaging 4.8 minutes apart. For each person 110 variables were available, including ability test scores, interests, and ratings of behaviour, all of which are supposed to be shown in the birth chart. The test conditions could hardly have been more conducive to success but the results were uniformly negative. The effect size due to astrology, expressed as a correlation on a scale of 0 to 1, was 0.00 ± 0.03.

The above result is consistent with other empirical studies, which when free of artifacts have consistently failed to find effects commensurate with astrological claims. Here, however, such a result is actually good news, because if artifact-free tests of astrologers are found to give positive results it might suggest the existence of human abilities of interest to parapsychologists.

Studies of Astrologer Accuracy

Tests of astrologer accuracy generally involve astrologers matching birth charts with information such as personality profiles or case histories, where a hit rate significantly better than chance would be evidence that they are on to something. Meta-analysis of more than forty controlled studies suggests that astrologers are unable to perform significantly better than chance even on the more basic tasks such as predicting extraversion. The mean effect size is 0.05 ± 0.12, p = 0.66, equivalent to getting 52.5% hits when 50% is expected by chance. Visual plots indicate the existence of a publication bias against negative results, which probably accounts for the weak positive direction.

More specifically, astrologers who claim to use psychic ability perform no better than those who do not. The results are no better when the astrologers are given all the case history material they ask for including photographs. Nor can people pick their own chart reading when given several to choose from, a result based on ten studies in which give-away cues were known to be absent.

There is clearly nothing here to suggest that astrologers can perform usefully better than chance, or that your own chart reading fits you better than someone else’s, regardless of how the reading is made or how well the conditions are stacked in the astrologer’s favour.

Studies of Astrologer Agreement

Astrologers tend to dismiss the above findings as the result of asking the wrong questions, albeit without ever telling us what the right questions should be. But such ploys are irrelevant when testing agreement among astrologers, simply because the truth of their answers and the usual philosophical arguments about truth are now of no consequence. Our only concern is agreement, the extent to which astrologers agree on what a given chart indicates.

Meta-analysis of 25 controlled studies suggests that astrologers show negligible agreement on their chartt readings, and equally poor agreement on their confidence in those readings. The point is, we might conceivably explain away poor effect sizes for hits, but not poor agreement or the inconsequence of confidence. If astrologers cannot agree on what a given birth chart indicates, or on their confidence in that indication, then what price astrology and the supposed intuitions of astrologers?

Conclusion

Our concern in this article has been to measure the performance of astrology and astrologers. A large-scale test of time twins involving more than 100 cognitive, behavioural, physical and other variables found no hint of support for the claims of astrology. Consequently, if astrologers could perform better than chance, this might support their claim that reading specifics from birth charts depends on psychic ability and a transcendent reality related to consciousness. But tests incomparably more powerful than those available to the ancients have failed to find effect sizes beyond those due to non-astrological factors such as statistical artifacts and inferential biases. The possibility that astrology might be relevant to consciousness and psi is not denied, but if psychic influences exist in astrology, they would seem to be very weak or very rare. Support for psychic claims seems unlikely.

Further reading

For critical articles on astrology see www.astrology-and-science.com/

Postscript for Butterflies and Wheels

The JCS article was featured in a half-page article in the UK Sunday Telegraph 17 August 2003 that raised a storm of protest from astrologers who generally showed no awareness of relevant research or even of the original JCS article. An example is the article by astrologer Neil Spencer in The Guardian 19 August 2003, which featured briefly on the B&W website under the heading “Oh That Old Ploy.” The objections he raised were already fully answered by either the JCS article or previous publications, and thus made no sense. (Spencer’s article is at www.guardian.co.uk/g2/story/0,3604,1021328,00.html)

More subtly, the protesting astrologers saw the JCS article as the case against astrology when it was merely an examination of astrology’s relevance to consciousness and psi. Our other articles on astrology have included the case for astrology, which is summarised in the following excerpt from www.astrology-and-science.com, starting with the case against:

The case against astrology

The case against astrology is that it is untrue. It does not deliver benefits beyond those produced by non-astrological factors, it has not contributed to human knowledge, it has no acceptable mechanism, its principles are invalid, and it has failed hundreds of tests. But no hint of these problems will be found in astrology books, which in effect are exercises in deception. But it doesn’t end there.

Astrologers disagree on almost everything, even on basics such as which zodiac to use. They rarely test control data, which is why scientists see astrologers as crazy or even crooks. In fact astrologers are mostly nice people who genuinely wish to help others. But the claim they repeatedly make (astrology is true because based on experience) is simply mistaken – what they see as its strength (experience) is actually its weakness (no controls).

The case for astrology

The case for astrology is that a warm and sympathetic astrologer provides low-cost non-threatening therapy that is otherwise hard to come by. You get emotional comfort, spiritual support, and interesting ideas to stimulate self-examination. In a dehumanised society astrology provides ego support at a very low price. Where else can you get this sort of thing these days?

In short, there is more to astrology than being true or false. But note the dilemma – to get the benefits you have to believe in something that is untrue. The same dilemma can apply elsewhere as in psychotherapy and even religion, so it is not unique to astrology. Nevertheless it presents an ethical problem that astrologers have generally failed to recognise let alone resolve.