From www.astrology-and-science.com       12m 12g 318kb       Home       Fast-Find Index

That was 5 decades ago. Since then there have been new tests, new approaches, and a thousand new studies using computers for astrological calculations previously impossible by hand. In other words a New Understanding is now within reach. But books continue to parade views either for or against based on incomplete surveys of a mostly hard-to-find literature with no nuanced views. So this new survey collects all the evidence -- not just the bits selected to prove a point -- and subjects it to rigorous critical thinking. The results explain why for many people there is more to astrology than being true or false.

Ward's earlier review was quite thorough (1480 words). He quotes Recent Advances: "The astrological literature is filled largely with demonstrations of belief. What it is not filled with is demonstrations of truth;" and sees Tests of Astrology as supporting and strengthening this assertion: "It includes personal stories of astrologers and believers who changed their minds; the discovery of astrology; the evolution of tests and why they are needed; a whole section on the Michel Gauquelin work; hundreds of empirical studies over 1927-2015; test overviews; artifacts (including a good general discussion of why people see what they expect to see); the future of astrology; the case for and against; a glossary; and name, subject and book indexes."

He concludes with his own overview: [Tests of Astrology] ends with a summary prefixed by a quotation from Christopher Hitchens 'What can be asserted without proof can be dismissed without proof': there is no physical way astrology could work; hundreds of tests have shown it does not deliver useful factual truth; the claimed as above, so below links do not exist; charts are meaningful even when wrong; outcomes are explained by hidden persuaders; and claimed experience is unfounded because it is never tested under controlled conditions. Hits are chosen, misses ignored, failures explained away, and unwelcome test results dismissed because, it is said, astrology cannot be tested.

Understanding Astrology is a thorough compilation and review of "under the radar" [ie hard-to-find] research, much of which is quite interesting. Its text is divided into eleven chapters. After the introduction, Chapter 2, Personal Stories, is unusual, entertaining, and informative. Chapter 3 is a brief (fourteen-page) history of astrology, including the Chinese and Hindu versions. Chapter 4, Disagreements in Astrology, covers both disagreements among astrologers and between astrologers and scientists.

Chapter 5 describes the early (starting in the 1500s) tests of astrology. Chapter 6 is devoted to one of the most contentious controversies in modern astrological research, Michel Gauquelin's [Mars effect]. The authors argue that the effect is real but not due to astrological influences. Their argument is both complex and convincing.

Chapter 7, Individual Studies, is by far the longest chapter in the book at 556 pages, or 60 percent of the total page count. Studies are described along with critiques when the authors of the studies claimed that their results showed a relationship when they did not. This an especially useful characteristic of the book. Studies are not just described but are critiqued. Where authors have made logical, methodological, or statistical errors, these are noted. Some of the studies summarized, as well as some of the authors' critiques, use sophisticated statistical analyses.

While Chapter 7 described and discussed each test of astrology individually, Chapter 8, Overviews, discusses tests by topic, such as tests of sun signs, tests of matching horoscopes to personality test results, and even tests of [time twins]. The chapter also covers statistical issues and concludes that there is no evidence that astrology has any validity. Those few tests that seem to support astrological predictions can "be explained by knowledge of serious astrology biasing the results." The authors (p.862) predict that "astrologers will no doubt see this conclusion as confirming the inability of critics to open their closed mind". Chapter 9 concerns artifacts that lead to belief in astrology, such as reliance on personal experience, confirmation bias, seeing patterns in random noise, and the representativeness heuristic, among others.

Chapter 10, Future of Astrology, is a short chapter of sixteen pages. This is appropriate because the large corpus of studies reviewed in this book show that astrology has no empirical foundation. Some astrologers resort to using it as a basis for counseling. But the authors wisely note that "Astrology by itself is not counselling. People with problems need to learn coping skills, but this will not happen unless the astrologer is properly qualified." Chapter 11, The Case for and against Astrology, approaches the issue from a broader perspective that asks whether astrology can be seen as a belief system that can function and be beneficial even though it is scientifically invalid.

To say that Understanding Astrology is comprehensive is an understatement. It covers every aspect of Western astrology that I can imagine in clear and concise detail. It took a huge amount of work to compile the information in the book. The coverage is not limited to material in English. The text is extensively illustrated with reproductions of diagrams and figures from the publications discussed. Also delightfully included are copies of relevant book covers, artwork, and amusing cartoons. In short, it is the definitive work on the status of astrology.

First in his editorial:

"[The authors] make no substantial criticism of important recent research. ... They misreport data and they argue by using strawman and other fallacies. There are no discoveries of artefacts, no issue of lack of control groups, no reporting of statistical errors or under-sized samples ... This unscientific approach is symptomatic of critics in decline who need an easier target than empiricism allows" (p.6).

Then in his review as follows:

"Instead of a critical review the authors present fallacies and misinformation. The thousand viable empirical studies do not exist and while the authors understand the techniques of astrology, they lack a genuine understanding of the practice of astrology" (pp.80-81).

Readers aware of Understanding Astrology's 4000 references might wonder how a thousand studies could suddenly disappear. They might also wonder how Currey could disagree so strongly not only with each of the red astrologers in the above list but also with each of the blue academics whose qualifications are world class, and with the serious reviewers when they note the detailed treatment of issues that Currey claims are ignored. For example Currey dismisses the discussion on Gauquelin's work as "far-fetched ... unsupported by evidence ... unfalsifiable" (p.65) whereas Professor Hines sees it as "complex and convincing".

Conflicts of interest
Arguably Currey's review reflects his many years of experience selling his astrological services and computerised chart readings. Until Dell Horoscope failed due to Covid, it often contained his full-page ads "Tell Robert Currey three things about yourself [date, place, and time of birth] and he'll tell you everything about yourself", which must inevitably raise concerns about conflicts of interest. Even if his income from astrology were suddenly zero, his making money from previous years of astrology would still remain a vested interest.

Indeed, Currey (p.72) says he spent years perfecting his chart readings. But so did Terry Dwyer to perfect his popular Starword system of chart interpretation, concluding "It gets good results by being thorough and systematic" (Astrological Journal 25(3), 1983, p.197). But when Dwyer later applied controls, his "good results" were no better than chance, see UA pp.792-793. Currey reports no controlled tests of his own system.

In sharp contrast, none of the four UA compilers make money from astrology or from sales of their astrology books. Even when two of them (Dean and Smit) were professional astrologers, they immediately quit when their controlled tests failed to confirm astrology's validity. That was 35 years ago. Nevertheless Currey questions their "potential conflicts of interest" because Understanding Astrology is distributed "without being commercially available" (p.73). Which makes no sense because their books have always been commercially available, and anyone can buy copies from AinO Publications for the cover price advertised on this website and elsewhere.

Avoidable errors
Aside from Currey's conflicts of interest, there are too many errors in Currey's quotes for his review to be taken seriously. Here are six examples. The first four are from p.69:

Currey refers to "the largest and least astrologically successful experiment [by] Narlikar (2009)". But that 2009 experiment was by Rajopadhye not Narlikar (see UA p.749).

Currey says "I salute Dean for being the first to have a go at meta-analysing astrological research". But that honour goes to the late Dr Michael Startup in his 1984 PhD thesis (see UA pp.232-233).

Currey complains about UA's use of Cohen's kappa for calculating effect size for which it is a "flawed metric" (his complaint is expanded to two pages in Correlation 34(2), 52-53, 2022). But the problems with kappa as an effect size are fully explored with examples in UA p.756, and kappa is used only where flaws do not occur.

Currey says "The meta-analyses are cherry-picked. [Those previously published] appear to have been cut from UA". But none have been cut. Some have been expanded. All can be found via the index.

Currey claims "The authors' approach to Effect Size suggests that they do not appreciate how astrology is practised ... they do not understand astrology" (p.72). Here UA pages 4 and 25 present the facts. Mather was Research Co-ordinator for the UK Astrological Association 1971-1978, editor of its 1970s research journal, and a consulting editor of Correlation since it began in 1981. Both Dean and Smit are former full-time practising astrologers and teachers of astrology. Dean was the first president of the Federation of Australian Astrologers in Perth. Smit founded the Dutch Society for Professional Astrologers and was its first chairman. He was also the founding editor of the journal of NVWOA, the Dutch Society for Scientific Study into Astrology, and for six years was the editor of Correlation. Nias was the co-author with Professor Hans Eysenck of Astrology: Science or Superstition?, which became the standard reference for academics and was translated into nine languages. Together with the publisher they have two doctorates and two masters in directly relevant areas.

Finally Currey p.66 quotes Mather's comment "I now believe [the future of astrological practice] lies in the direction of recognising astrology as a divination", citing UA p.913 and claims it shows an "apparent change of heart". But p.913 does not mention Mather. Mather's full quote appears on UA p.48, and shows that Currey's citing p.913 is seriously incomplete. It should have ended "as a divination as defined in 11.4.2". As it happens Section 11.4.2 does appear on p.913 and specifies divination "in the special sense of giving us new viewpoints from which to explore issues, not in the popular sense of fortune-telling". In other words an "apparent change of heart" cannot be read from the quoted pages.

Even worse than Currey's misquotes is his approach to what he calls "meta-analysis", where he uses a cherry-picked list of positive studies from Correlation to show how astrology gives positive test results. The problem is that his "meta-analysis" is without the unbiased sampling and secondary statistical analysis that a proper meta-analysis requires, both of which are detailed in UA where page 760 gives a basic procedure from Hunter & Schmidt's Methods of Meta-Analysis with equations and instructions sufficient for anyone to get started. When Currey's data is properly meta-analysed, the support for astrology disappears:

Above: A proper meta-analysis of Currey's 10 studies is not significant p=0.35. Nor does it differ from (1) a meta-analysis of 70 studies of astrologers' ability to match charts to subjects p=0.65, or (2) a test of 152 astrologers' ability to pick the correct chart out of 5 for a dozen subjects p=0.35 (it also found that world-class astrologers were no more successful than beginners). In his AA Kepler Day talk "Introduction to Statistics" in January 2025, Currey added 12 more studies to his "meta-analysis", again none of them involving astrologers reading charts, and claimed they increased the effect size r from 0.21 to 0.29. But a proper meta-analysis of the extra 12 gives r=0.048 ±0.056 p=0.42, much the same as the original 10, as was all 22 combined, namely r=0.048 ±0.056 p=0.39. Currey's "meta-analyses" were consistently misleading. What they do is precisely what authentic meta-analyses avoid.

For readers unfamiliar with meta-analysis, it weights a set of effect sizes by their sample sizes, subtracts the sampling error (something not possible with a single effect size), and checks the results statistically to see if there is a real effect. Care must be taken when selecting studies otherwise the results will not reflect reality. So, regardless of topic, the meta-analyst has to include all relevant studies whether published or unpublished, positive or negative, to avoid publication bias against negative results. Such bias is well-known and less common these days among orthodox journals but is usually automatic among astrology journals.

That said, it still makes good sense to include questionable studies in a preliminary meta-analysis to see what effect they have. Being informed is better than having to rely on guessing.

The good news is that since the 1970s meta-analysis has revolutionised research in the sciences and humanities by allowing clearer conclusions where once there was only uncertainty. The technique is detailed in dozens of books and by 2015 about 10,000 meta-analyses had been published. The bad news is that readers may well see Currey's improper use of such a well-established technique as a serious blow to the scientific credibility of astrological research in general and of Correlation in particular.

But enough is enough. By now it should be clear that a summary of Currey's review without correction of its errors and improper use of "meta-analysis" would be premature. If you disagree, you can avoid endless opinion wars by closely reading both Currey and UA (and where necessary the original literature) and then making up your own mind. More on this below (look for "free download" just before the heading "Not far-fetched").

All major astrological bodies have a constitution or mission statement that essentially reduces their objects to "To advance astrology", usually specifying the means such as holding classes. A few specify the study area, such as "correspondences and cycles" (NCGR), or "all scientific methods of astrology" (AFA), or simply "astrology in all its branches" (Astrological Lodge of London, founded in 1915, and like the AA a non-profit UK Charitable Incorporated Organisation).

In the new Constitution the objects of the AA are still to advance astrology by the study of astrology, but they differ from those of other astrological bodies in specifying not just "study" but also its quality, namely "the critical study of astrology in all its branches". The meaning of critical is not defined but is generally understood to involve critical thinking and precautions against the many ways of being fooled when drawing conclusions. Both of course are automatic in science (to which the AA is traditionally inclined) and scholarly writing in general. It means that a study without these qualities -- such as those that routinely occur in any astrology class -- could not be described as critical, and would therefore not conform to what the Constitution specifically requires.

Clause 4 confirms the AA's "power to do anything which is calculated to further its object[s]", ie to advance astrology by critically studying it. Ordinary study as in any astrology course clearly won't do.

Trust the trustees
The AA is currently organised by 11 trustees (they include Gillett) who are required by Clause 7 to avoid conflicts between (1) their duty to act in the AA's interests by submitting astrology to proper critical study, and (2) their personal and financial interests. "Any charity trustee absenting [themself] from any discussions in accordance with this clause must not vote or be counted as part of the quorum ... on the matter". In other words the Constitution does not allow any trustee to promote vested interests or to promote views not critical of astrology. So readers can confidently expect Gillett's review to meet the Constitution's requirements. But what do they actually get?

In what follows any unreferenced quote is from Chapters 1-3 (eBook pp.13-30) of Gillett's Working with the Planets, Crucial Books, Camberley 2021.

Gillett wastes no time in promoting his personal experience over hard evidence. His 930 words dismiss UA's thousand studies in favour of his own books. For example he finds that UA has no appreciation of "the valuable way genuine astro-cycle study can be used to understand daily experience" produced by "background pressure (outer-planetary cycles), the contemporary mood (inner planet cycles), and triggers to act (Moon and angles)". These cycles are not based on collective critical research but on his personal experience, which he describes in his book but not in his review (the book's cover shown above dominates his review but there is no visual of UA). So in fairness we should take a look at his personal experience:

The role of personal experience
Years of "thumbing the ephemeris ... scanning my birth chart and those of friends ... and travelling the world with astrology as my guide" had convinced Gillett that astrology was "useful as a diagnostic tool that measures pressures, possibilities and social trends". He gained "insight into the neat way astrology fits and reflects events exactly". Nevertheless modern astronomers "rip out the very soul of meaning from our lives", which means "we still need something that defines our world ... for ourselves, our family, and friends". In other words Gillett's personal experience is his arbiter of truth.

The key role played by Gillett's personal experience is confirmed by the rest of his book. It consists of annotated extracts from his "Working with the Planets" column in the Astrological Journal 2002-2021. They show how birth charts exactly match lives and events, which to Gillett is clearly better than any "modern [soulless] astronomical understanding of the heavens". Indeed, just "Check it out for yourself". Put like this, it seems hard to disagree. But there is a problem:

Wrong charts
It is a common experience among astrologers including Gillett that an exact match exists between a chart and its owner even when the chart is later found to be wrong (by hours, days, years, it makes no difference; UA gives nearly 30 examples see pages 856-858). In which case the match cannot be due to astrology. So Gillett's arguments in favour of astrology necessarily collapse. How does he respond to this?

Gillett's only comment is: "Have you ever been progressing really well through a session with a client only to realise the time of birth is wrong? Then there are mistakes with time zones and all sorts of other embarrassingly silly errors that you kick yourself for. Take heart! If the motivation for the work was free of ego, then you can be sure the mistake was made for a very good reason. Such mistakes, I prefer to call 'magical errors'."

Gillett seems to be implying that astrology works because of some kind of magic and not necessarily because of anything in the birth chart. In any case you need only try again with the correct chart to appreciate "the deeper understanding that follows".

Except this merely evades the problem (that a match between subject and chart means nothing if it can exist for the wrong chart). Fortunately the problem does not arise in one of the most common tests of astrologers where they have to pick the correct chart out of several (see UA p.786 ff). The latest was an independent test of 152 astrologers from beginners to world-class experts (S Greenberg et al, 2024, www.clearerthinking.org/post/ does-astrology-work?). They had to pick the correct chart out of five for 12 subjects with certified birth times and answers to 43 multiple-choice questions chosen by astrologers as being the most relevant to picking a person's birth chart. The first astrologer to get at least 11 charts right would win a prize of $1000 -- and after finishing the test 15% of the 152 were confident of success, especially if they were world-class.

In fact the world-class astrologers performed no better than the beginners. Indeed, not one of the 152 astrologers performed consistently better than chance despite the 15% who felt confident of success. Unless Gillett can do better, it is not clear how his personal experience can be seen as supporting astrology as a source of knowledge.

Unreliability of personal experience
Another problem evaded by Gillett is the unreliability of personal experience. For example centuries of bloodletting were accepted as valid due to personal experience, and the same was true for decades of phrenology and years of biorhythms (see UA p.639). All were kept alive by personal experience. But when tested against controls that kept conditions the same except for deliberately changing a key component such as the subject, personal experience was shown to be irrelevant to the truth. So the beliefs necessarily collapsed. It made no difference that the beliefs had once seemed to rule the world.

If you are unconvinced, note that UA compilers Dean and Smit were just as dedicated to astrology as is Gillett. Like Gillett, both were successful practitioners, and both were head of their respective large astrology groups. But unlike Gillett their science training led them to make controlled tests. When the results contradicted their dedicated beliefs they immediately gave up reading charts for clients and making money from astrology since it would have been as unethical as claiming bloodletting, phrenology and biorhythms still worked. They had discovered the hard way that personal experience without careful empirical support could be massively misleading. Despite the best of intentions, personal experience was too easily fooled.

In other words, to meet the AA's Constitutional requirement for critical study, AA trustees like Gillett need the support of controlled tests and replications. But Gillett cites no such support other than a vague reference (his review p.29) to Currey's "meta-analysis" in Correlation 34.2. Instead he tends to rely on general statements that are too unclear to be directly addressed (for example he complains p.28 that UA is "lacking a proper model of astrology", but without telling us what a proper model is); and on appeals to general claims such as p.29 an "indulgent desire for money (Pluto in Scorpio)", with no concern for how controlled tests have consistently failed to validate signs, see UA p.772.

He also makes too many mistakes for his review to be taken seriously. For example he dismisses "How Astrology Works" (UA pp.89-92) as "just five pages of disconnected opinions" (his p.29). But the correct title is "Disagreement on how astrology works". So what is he talking about?

Answering astrology's critics
Nevertheless he does consider the problem of "Answering Astrology's Critics" (the quotes that follow are from his eBook pages 613-622). According to Gillett, the reason academia ignores astrology as a source of knowledge is "ill-informed simplistic ignorance". Similarly, "to become acceptable, astrology does not need to prove itself before ... the modern material scientific view". UK physicist Professor Brian Cox "went to extreme lengths to demonstrate his ignorance of astrology and prejudice against it". When our present society is viewed from the future, we "may well see our scientific focus as narrow and doctrinaire". To experiment with astrology "we should not seek to prove specifics, but use specifics to extend understanding". But how does this work? He gives no examples, but seems to be proposing the use of accepted beliefs without question.

Gillett feels that some specifics "like those studied in Gauquelin's work, are so strong they look like mechanical proofs". That they also disagree with traditional ideas (eg no effect for the Sun, see UA pp.159-161) is somehow of no consequence. Again, "If people listened to ... experienced astrologers, the world and our lives would be much happier" (how can he know this?). And so on. Exactly how this parade of opinions provides an answer to serious questions (such as why identical and fraternal twins who share almost the same birth moment are not more alike than ordinary siblings who share very different birth moments), is not clear.

Indeed, rather than addressing the problem of vested interests by quoting scholarly articles by others, Gillett makes it worse by mostly quoting his personal views, claiming that Understanding Astrology has only "tedious pages ... failed studies ... nit-picking narratives ... based on sand ... disconnected opinions ... and authors who misunderstand astrology and what astrologers do" (recall that Currey wrongly claimed the same). So readers who might be interested in a thousand studies, some by world-class astrologers, needn't bother because Gillett prefers his own books (his are the only books in his reference list, even UA does not appear).

But why listen to Gillett when you can ask Google if there is "any science to back up whether astrology impacts our personality and our lives?" and (since June 2020) receive "Here's the short Answer: No. None whatsoever". On the other hand, why take Google's word for it when what you need is a combined study by qualified astrologers and qualified scientists with full attention to detail, due concern for the relevant literature, and of course no vested interests. Such studies used to be difficult if not impossible to find except perhaps in Recent Advances (see above following our abstract) but today there is Understanding Astrology available both as a free download of nearly 1000 pages and as a convenient lie-flat book for the price of ten takeaway cups of coffee.

Understanding Astrology: A critical review of a thousand empirical studies 1900-2020 by Geoffrey Dean, Arthur Mather, David Nias, and Rudolf Smit, published in 2022 by AinO Publications in Amsterdam. This massive volume (952 pages) is a labour of love by four of the world's leading experts on empirical tests of astrology and has taken many decades to compile. It really does tell you everything you ever wanted to know about astrology.

Professional astrologers are often very dismissive of sun sign astrology. They insist that the power of astrology can only be revealed by examination of "the real thing", that is, the casting of a full horoscope by a professional astrologer based on exact birth details which is then interpreted in a face-to-face consultation between astrologer and client. The truth is that, when properly tested [ie with proper controls] the "real thing" turns out to have exactly the same level of validity as sun sign astrology, that is to say, none whatsoever.

When I was running my anomalistic psychology module at Goldsmiths I used to set the following essay title as a tutorial topic: Does astrology work? The best essays were those which essentially answered "Yes and No". So in what sense does astrology work? Dean et al (p.911) summarise the case for astrology: "A warm and sympathetic astrologer can provide non-threatening 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."

But when it comes to the question of whether astrology works in the sense of having any scientific validity, the answer is less encouraging [simply because] the predictions of astrology have now been put to the empirical test (at least) a thousand times. The results overwhelmingly fail to support the scientific validity of astrology. When occasionally apparently positive significant results in support of astrology are obtained, they are invariably found to be due to such factors as inappropriate statistical analysis, inadequate sample sizes, and a range of other artifacts all meticulously detailed.

Chapter 7 (558 pages) provides critical analysis of individual studies of astrology's claims. Chapter 8 (112 pages) helpfully provides overviews of tests divided up into a number of different categories. With respect to signs, the conclusion is clear: "signs (not just sun signs) [are] the most tested and most disconfirmed idea in astrology" (p.772). When it comes to tests of astrologers themselves, the picture is equally bleak. For one thing, the degree of agreement between astrologers assessing the same chart is abysmally low. Tests relating to geophysical factors, time twins, predictions, horary astrology, mind-related factors, divination, and wrong charts fare no better. The overall conclusion is inescapable. Despite a huge amount of time, effort, and resources having been directed at testing astrology, there is no evidence whatsoever that it has any useful scientific validity.

[Which leaves the obvious question] why do so many people believe in it? Dean et al do an excellent job of addressing this question in Chapter 9 [Artifacts, 22 pages] in terms of such factors as the Barnum effect, cold reading, subjective validation, making the chart fit the client, making the client fit the chart, selective memory, self-fulfilling prophecies, and the use of (generally) positive readings. When astrology is assessed in terms of "does it work?", I think any reasonable person would agree with the best essays in my tutorial on astrology. Or as Dean et al put it (pages 3 and 907), "there is more to astrology than being true or false".