Astrology, data, and statistics

Is western astrology a big data science, or even the very first big data science? Data scientist Alexander Boxer thinks it is and has written a book to back up his claim, A Scheme of HeavenThe History of Astrology and The Search for Our Destiny in Data.[1] 

His justification for having written this book is interesting:

Over two thousand years ago, astrologers became the first to stumble upon the powerful storytelling possibilities inherent in numerical data, possibilities that become all the more persuasive when presented graphically in a chart or figure. Although it took a while for the rest of the world to catch on, the art of weaving a story out of numbers of figures, often a specific course of action, is used everywhere today, from financial forecasts to dieting advice to weather models.

And yet numbers still mislead, figures still mislead, figures still deceive, and predictions still fail–sometimes spectacularly so–even those that rely on exceptionally sophisticated mathematics. So, are the techniques being used today to parse and package quantitative information any more effective that what was devised by astrologers millennia ago?

            In order to make that assessment, it’s first necessary to have a basic understanding of what astrology is and how it works. But that sort of understanding–one that’s at least adequate to resolve some seemingly straightforward technical questions–is surprisingly hard to come by for such a long-lived and influential craft. Being frustrated in my own search for a simple yet competent overview of astrology, I decided I might just as well write one myself. This, curious reader, is the book you now hold in your hands.

Boxer is actually correct “a simple yet competent overview of astrology” doesn’t, as far as I know, exist, so has he succeeded in providing one? My answer is a qualified “yes, no, maybe, probably not!” Large parts of Boxer’s book are excellent, other parts are OK, some parts I found simply baffling, and one of his central claims is simply wrong. The biggest problem with the book, as far as I’m concerned, is that it tries to be too many different things in far too few pages. It wants to be a history of astrology from its beginnings down to the present days, at the same time being a data scientist’s, statistical analysis of fundamental aspects of astrology, as well as presenting a quasi-philosophy of science meta-analysis of some central themes of astrology, and that all whilst attempting to achieve to authors declared central aim of providing “a simple yet competent overview of [western][2] astrology.” All of this in just 263 pages of an octavo book with a medium typeface. He also largely leaves out any serious attempt to present the interpretation of a horoscope, which is actually the essence of astrology.

The excellent bits of Boxer’s book are almost all confined to the technical and mathematical aspects of casting a horoscope and to the data scientist’s statistical analysis of various aspects of astrology. There is for example a competently presented, entire chapter devoted to the nuts and bolts of mathematical astronomy, without which it is impossible to actually cast a horoscope. However, this illustrates one, in my opinion serious error in the book. In the opening chapter Boxer presents a brief greatest hits tour of what he labels the obscure beginnings of astrology. I’ve read accounts of the material he presents here that are longer than his entire book, to which I’ll return in a minute, but that is not what concerns me at the moment. Here he presents for the second time (the first one in in the introduction) one of the excellent illustrations that occur throughout the book. This is a horoscope presented on the mater and tympan of an astrolabe without the rete but with the ecliptic. Also presented are all of the relevant astronomical data, time, in various formats, celestial coordinates in all three variants, geographical coordinates and so forth. See below:

However, there is absolutely no explanation of what is being presented here. Now, I’ve spent a number of years studying this stuff, so I know roughly what I’m looking at, although I need to look up which celestial coordinate system is which, for example. A naïve reader coming to this book to learn about astrology would have no idea what they are looking at and nowhere in the book do they get this diagram explained carefully step for step. The knowledge required is contained in the book, scattered around in various sections and chapters but with no linking references to the diagrams. The celestial coordinates are, for example, explained in the chapter on mathematical astronomy, whereas the astrolabe only gets explained in dribs and drabs about one hundred pages later in the book. The Julian Day Count, one of the methods listed on the diagram to denote the time of the horoscope only gets explained on pages 225-226! The information needed to understand what is in fact an excellent diagram is scattered throughout the book like a scavenger hunt without rules or clues.

Remaining by the topic, the book is liberally illustrated with diagrams and tables to explain themes under discussion, and these are excellently done both from a pedagogical and a graphical viewpoint and this is one of the great strengths of the book. There is not a conventional bibliography but at the end of the book there is an annotated collection of source material for each section of the book. There is also a competent index. 

Following up on the all too brief sketch of the origins of western astrology and the more comprehensive introduction to the basics of astronomy, Boxer now dives into what is without doubt one of the greatest error in the book, he fell in love with Marcus Menilius’ Astronomica. After briefly dismissing our knowledge of astronomy in the last five centuries BCE, a serious error because we actually know far more that Boxer is prepared to admit. However, if he did acknowledge it, he would have to abandon his love affair with Manilius. Boxer correctly explains that although the Roman took over large parts of Alexander’s Hellenistic Empire, they were initially reluctant to adopt the Hellenistic astrology. He illustrates this with the fact that there are absolutely no astrological discussions of Julius Caesar’s assassination in 44 BCE. Enter Marcus Manilius and his Astronomica stage left. 

A brief explanation, the Astronomica is a Latin didactic poem dating to the early first century CE, which happens to be the earliest surviving, relatively complete account of western astrology.  About its probable author Marcus Manilius, we know next to nothing. 

Boxer goes complexly overboard about the Astronomica. He writes:

The Astronomica is a fascinating work in its own right, but it takes on a special significance when we recognise that this poem is, essentially, astrology’s grand unveiling on the historical stage. And like Minerva issuing from Jupiter’s skull fully grown and clad in armour, the Astronomica presents an astrology emerging from obscurity remarkably complete and fully formed. Even today, two thousand years later, there is hardly any astrological idea, no matter how sophisticated or complex, which can’t trace its debut to Manilius’s poem.

If the Astronomica is “astrology’s grand unveiling on the historical stage” then it must have got lousy reviews from the critics. Not one single author in antiquity is known to have quoted the Astronomica. There are a grand total of about thirty existing medieval manuscripts of the work none of them older than the ninth century CE. It does not feature in any other medieval literature and appears to have been largely ignored in the Middle Ages. It was (re)discovered in c. 1416 by the zealous Renaissance Humanist manuscript hunter, Poggio Bracciolini (1380–1459) and only really emerged on the European literary and scientific stage when the editio princeps was published by Regiomontanus (1436–1476) in Nürnberg in 1473. 

In his love affair with the Astronomica, Boxer seems to think that modern horoscope astrology is somehow a Roman invention. Later in the book when taking about Arabic astrology he describes Masha’allah’s theory of astrological historical cycles as the “most significant addition to astrology since Roman times.” Manilius is in fact merely describing an existing system that was created by the Hellenistic Greeks between the fifth and first centuries BCE, something that Boxer acknowledges elsewhere in his book, when he goes overboard about the wonders of ancient Alexandria.

As for the guff about “astrology emerging from obscurity remarkably complete and fully formed” and “there is hardly any astrological idea, no matter how sophisticated or complex, which can’t trace its debut to Manilius’s poem,” as already stated Manilius is reporting on an existing system not creating it. More importantly as the modern commentators point out you wouldn’t be able to cast a horoscope having read it and it contains nothing on planetary influence in astrology, the very heart of the discipline.  In fact, although they adopted astrology and used it widely until the decline of the Empire, in the sixth century, the Romans actually contributed next to nothing to the history of astrology.

However, the chapter ends with an example of Boxer’s biggest strength the data based statistical analysis of various aspect of astrology. He starts here with the personality traits that Manlius attributes to those born under a particular sun sign, setting them out in a handy table first. Using the data of different professional groups, he introduces the reader to the concept of statistical significance and shows that the astrological divisions into personality types doesn’t hold water.

Next up we have Ptolemy the most significant author in the whole of the history of western astrology. He gives an adequate sketch of Ptolemy’s contributions to astronomy, geography and astrology and shows that they are actually three aspects of one intellectual project. In his brief discussion of map projection, he makes not an error, but a misleading statement. Introducing Ptolemy’s Planisphere and the stereographic projection the key to the astrolabe he writes:

For the basic idea of a stereographic projection, imagine looking down on a globe from above its North Pole [my emphasis], and then squashing in into the equator. The visual effect ends up looking like a scoop of ice cream that’s melted onto a warm plate from the bottom out. Because there’s no limit to how far outward these maps spread, it’s customary to extend them only as far as the Tropic of Capricorn.

The following pages contain stereographic projections of the celestial sphere, the terrestrial sphere and four tympans from astrolabes taken for different latitudes. Boxer’s error is that these are taken from the South Pole as projection point. Almost all astrolabes are for the Northern Hemisphere and are projections from the South Pole, there are only a handful of Southern Hemisphere astrolabes with the North Pole as projection point. 

Boxer also makes an error in his etymology of the Name Almagest for Ptolemy’s Mathēmatikē Syntaxis. Almagest comes from the Arabic al-majistī, which in turn comes from the Greek megiste all of which mean the greatest. Boxer justifies this as follows:

The Almagest was the greatest of all ancient treatises on astronomy, just as Ptolemy was the greatest of ancient astronomers.

In fact, all of this derives from the alternative Greek name of the Mathēmatikē SyntaxisHē Megalē Syntaxis meaning The Great Treatise as opposed to a smaller work by Ptolemy on astronomy known as The Small Treatise. In other words, the Almagest is the big book on astronomy as opposed to the small book on astronomy.

Boxer has a rather negative opinion of Ptolemy’s Apotelesmatika commonly called the Tetrabiblos in Greek, or Quadripartitum in Latin, meaning four books, his big book on astrology. He finds it dry, technical, and uninspiring, unlike the Astronomica. After introducing Ptolemy’s astrological geography Boxer once again applies his statistical analysis to Ptolemy’s claims on the geographical acceptance of homosexuality comparing it with the modern data on the topic.

Boxer’s next target is the only substantial collection of actual horoscopes from antiquity, by the second century Hellenistic astrologer, Vettius Valens’ Anthologies. We move from the theoretical, Ptolemy, to the practical, Valens. Here Boxer once again reverts to his role as data scientist and gives an interesting seminar on the theme of “how unique is a horoscope? Along the way he sings a brief eulogy for ancient Alexandria as a centre for the mathematical sciences including of course astrology. He also makes a brief excursion into the philosophy of science evoking the falsifiability criterion of Karl Popper and the separation of science and pseudoscience, a couple of pages that are far too brief for what is a very complex discussion and could have been happily edited out. His work, however, on codifying the basics of a horoscope according to Valens and examining the uniqueness of the result is stimulating and a high point of the book.

Next, Boxer moves onto medieval Arabic astrology but doesn’t really. He starts, as do many authors on this topic, with the horoscopes cast to determine the right time to found the city of Baghdad and having given a brief but largely correct account of why the Abbasid caliphs adopted astrology, and the parallel transmission of astrology into Europe in the High Middle Ages, he then passes rapidly to Masha’allah’s theory of historical cycles based on the conjunctions of Jupiter and Saturn and that’s it! Arabic astrology is a massive topic and given its powerful influence on astrology as its practiced today deserves much more attention in any book claiming to provide a “simple yet competent overview of astrology.” Once again, the chapters strength lies in Boxer’s statistics-based analysis of Masha’allah’s theory, which drifts off into the theories of encryption. One thing that did piss me off was in a discussion of the use of symbols he writes:

By necessity, then, efficacy of this magic will hinge upon the fitness of these symbols to their task: Nowhere is this more evident than in mathematics. (If you don’t believe me, try adding the Roman numerals CXXXIX and DCXXIII together; or, even worse, the Greek numerals 𝛒𝛌𝛉 and 𝛘𝛋𝛄.)

This is pure bullshit! Assuming that you are cognisant with the numeral systems and the values of the symbols than these additions are no more difficult than carrying out the same sums using Hindu-Arabic numerals. Division and multiplication are, at least at first glance, more difficult but there are algorithms for both numerical systems that also make those operations as easy as the algorithms for Hindu-Arabic numerals. The major point, however, is that nobody bothered; arithmetical calculations were carried out using an abacus and the numerals were only used to write down the results. 

Having very inadequately dealt with Arabic astrology, Boxer now turns to Guido Bonatti (died around 1300). Before he gets to him, we get a brief section on the transmission from Arabic into Latin where Boxer manages to conflate and confuse two periods of translation in Toledo, one of the major centres for that work. In the twelfth century translators such as Gerard of Cremona translated the major Greek scientific works from Arabic into Latin often with the help of Jewish intermediaries. Later in the thirteenth century Alfonso X of Castille set up a school of translators in Toledo translating Hebrew and Arabic texts into Latin and Castilian, establishing Castilian as a language of learning.  Boxer goes off into an unfounded speculation about texts being translated from Greek into Syriac into Arabic into Hebrew into Castilian (here Boxer incorrectly uses the term Spanish, a language that didn’t exist at the time) into Latin, with all the resulting errors. This paragraph should have been thrown out by a good editor. We then get a couple of paragraphs of waffle about the medieval universities that appears to exist purely to point out that Abelard and Héloïse named their son astrolabe. These should have been replaced with a sensible account of the medieval universities or thrown out by the same good editor. 

We then get an account of the twelfth and thirteenth centuries war between the Guelphs and Ghibellines in Northern Italy largely to introduce Guido Bonatti, who was a Guelph astrologer and author of the Liber Astronomiae, which Boxer tells us, hyperbolically, is the most influential astrology book of the Middle Ages. Here Boxer makes two major errors. Firstly, he presents judicial astrology, which he defines as follows:

The basic premise of judicial astrology is that you ask the stars a question–a question about pretty much anything–and the stars then reveal a judgement or, in Latin, iudicium. The astrologer’s job is to interpret these judgements on your behalf. So far, so good. The odd thing about judicial astrology, however, was that for many questions, and especially the broad category of yes-or-no questions, the astrologer would determine the stars’ judgement based on their positions in the sky at the moment your question was asked.

What Boxer is actually describing is horary astrology, just one of the four branches of judicial astrology, the other three are natal astrology, mundane astrology, and elective astrology; Boxer goes on later to discuss elective astrology. Judicial astrology was opposed to natural astrology, which meant astrometeorology and astromedicine, or to give it its proper name iatromathematics, neither of which Boxer deals with, in any depth, just giving a two-line nod to astromedicine. 

Having described horary astrology, albeit under the wrong label, Boxer goes off on a rant how ridiculous it is/was. Then come two more misleading statements, he writes:

Yet however ho-hum this fatalistic outlook may have been during astrology’s early days in Stoic Rome, to deny the existence of free will was a decidedly and damnably heretical opinion in medieval Christian Europe.

[…]

As was obvious to Dante. Petrarch, and many others, astrology–and especially judicial astrology–was fundamentally incompatible with Christian doctrine. 

First off, Stoic Rome was not astrology’s early days, by that time Hellenistic astrology had been around for about four to five hundred years. Yes, Hellenistic astrology was totally deterministic and did in fact clash with the Church doctrine of free will in the beginnings of the High Middle Ages. However, Albertus Magnus and Thomas Aquinas, who laid the foundations of Church doctrine down to the present day, redefined astrology in their writings in the thirteenth century, as acceptable but non-deterministic thus removing the doctrinal clash. In terms of the impact of their work for the acceptance of astrology not just in the Middle Ages, surely it is far more influential than Bonatti’s Liber Astronomiae.

In the passage that I left out of the quote above Boxer writes, amongst other things:

Well, that’s the sort of thinking that could get you burnt at the stake in you insisted on making a fuss about it. The astrologer Cecco d’Ascoli was condemned by the Inquisition on precisely these grounds and burnt at the stake in Florence on September 16, 1327. [i.e., for practicing deterministic astrology]

This is simply not true! In 1324, Cecco d’Ascoli was admonished by the Church and punished for his commentary on the Sphere of John de Sacrobosco, nothing whatsoever to do with astrology. To avoid his punishment he fled from Bologna, where he was professor for astrology, to Florence. Here, he was condemned for trying to determine the nativity of Christ by reading his horoscope, and as a repeat offender was burnt by the Inquisition. Even under the non-deterministic interpretation of judicial astrology from Albertus Magnus and Thomas Aquinas, casting the horoscope of Christ was considered unacceptable. 

Next, Boxer introduces the Houses of Heaven and claims that, “these are astrology’s system of local coordinates the astrological analog to the modern-day quantities azimuth an elevation.” Sorry but this statement is garbage the houses are not a coordinate system, they are divisions of the ecliptic plane. Boxer introduces them here because they play a central role in Bonatti’s horary astrology. Once again Boxer the data scientist comes to the fore with the question whether it would be possible to construct an algorithm to automatically answer questions posed in horary astrology. As usually one of the best parts of the book.

Traditionally, one of the major disputes amongst astrologers in the question how exactly to determine the boundaries of the houses and Boxer now turns his attention to the various solutions presenting nine different solutions that have been used at some time in the history of astrology. 

One system that was very popular in the Renaissance and Early Modern Period was devised by Regiomontanus (1436–1476), which Boxer looks at in somewhat more detail. He starts with a very brief rather hagiographical biographical sketch, which includes the following claim:

By the time he was twenty-six, Regiomontanus had finished a complete reworking Ptolemy’s Almagest using all the newest trigonometrical methods. 

The Epitome of the Almagest was commissioned from Georg von Peuerbach, Regiomontanus’ teacher, and later colleague, by Cardinal Basilios Bessarion in 1460. Peuerbach had only completed six of the thirteen books by 1461 when he died. On his death bed he commissioned Regiomontanus to complete the work. Regiomontanus went off to Italy with Bessarion, basically as his librarian, and spent the next four years travelling through Italy collecting and copying manuscripts for Bessarion’s library. During this time, he probably completed the Epitome. Meaning he was twenty-nine. Although he might have finished it during the next two years, when we don’t know where he was or what he was doing. He intended to publish the finished book when he set up his publishing house in Nürnberg in 1471 but still hadn’t by the time he died in 1476. It was first published by Johannes Hamman in Venice in 1496

Further on Boxer writes:

Thus, when a certain archbishop in Hungary demanded an improved system for determining the Houses of Heaven–in particular one that would be more faithful to the vague instructions given by Ptolemy in his Tetrabiblos–there was only one person to ask.

            Regiomontanus accepted the challenge. In a brash and masterly treatise, he surveyed the existing methods of House division, dismissed them all as inadequate, introduced an entire new method, and provided tables for computing their boundaries at any latitude to the nearest minute of arc.

A nice story but unfortunately not exactly true. The title of the book that Regiomontanus wrote at the request, not demand, of János Vitéz Archbishop of Esztergom, for whom he had been working as a librarian since 1467 was his Tabulae directionum profectionumque. The purpose and content of the book is revealed in the title, this is not a book about the determination of the Houses, which are only secondary product of the book but about calculating directions, also called prorogratio or progression from the original Greek aphesis. A method to determine major events in the life of a horoscope subject including their death, described by Ptolemy in the Tetrabiblos, which was very popular in Renaissance astrology. 

This error by Boxer is rather bizarre because he describes the method of aphesis, albeit wrongly, whilst dealing with Manilius earlier in his book. Here he writes:

…a procedure … entailed identifying two key points on a birth horoscope: the “starter” and “destroyer.” As time elapsed from the moment of birth, the destroyer revolved along with the heavens towards the starters original position, all the while shooting evil rays at it. When the destroyer finally reached the starter, it was game over: death. The number of hours and minutes it took for the destroyer to reach the starter was then converted to the number of years and months the individual was expected to live.

A very colourful description but actually fundamentally wrong. First the astrologer has to determine the starter on the ecliptic, which is often the moment of birth but not necessarily. Then various destroyers are identified signalling major events in the life of the subjects not just their death, also on the ecliptic. Both points, started and destroyer are projected using spherical trigonometry onto the celestial equator and the number of degrees between the projected points is the time in years. Regiomontanus’ Tabulae directionum provide the mathematical apparatus to carry out this not particularly simple mathematical process. 

Which system of Houses division is still disputed amongst astrologers and Boxer possesses the impertinence to suggest they should use a particular system because he finds it mathematically the most elegant. 

The chapter closes with a short discourse on time, unequal hours, and equinoctial hours, which serves two functions to introduce the index or rule on the astrolabe which makes possible the conversion between unequal and equal hours. Boxer then states:

That the development of the mechanical clock occurred precisely when the most intricate astrological algorithms were in vogue is a historical synchronicity too striking to ignore.

[…]

In fact, the technological crossover between astrology and clock design was significant.

Here he is referring back to an earlier statement on the previous page:

This is why the earliest mechanical clocks of which the one in Prague’s old town square is the most magnificent example had astrolabe-style faces.

Source: Wikimedia Commons

Unfortunately for Boxer’s enthusiasm David S Landes, a leading historian of the clock, argues convincingly that the simple mechanical clock with a “normal” clock face preceded the astrolabe-style clock faces.

The next chapter opens with Tycho Brahe and the nova of 1572. Here once again Boxer choses to distort history for dramatic effect. He writes:

Yet, by all accounts, Tycho wanted nothing to do with Denmark’s administration, its wars, its politics, or its pageantry.

            For a nobleman like Tycho, the purpose of a university education was not to obtain a degree–that would have been unthinkably déclassé–but merely to pick up a little worldly polish of the sort that might prove serviceable in war and diplomacy. In this respect, Tycho’s education backfired spectacularly. He returned from Germany utterly captivated by the latest advances in alchemy, astronomy, and astrology.

Boxer carries on in this manner presenting Tycho as a rebel kicking against the pricks. What he neglects to mention is that although Tycho’s decision to become a professional astronomer was somewhat unorthodox, in all his endeavours Tycho received strong support from his maternal uncle Peder Oxe. Oxe was a university graduate, and a strong supporter of Paracelsian alchemical medicine, who just happened to be the Danish finance minister and Steward of the Realm, de facto prime minister, and politically by far the most powerful man in the whole of Denmark. 

Boxer closes his short section on Tycho with another piece of purple prose:

Tycho’s supernova is of tremendous historical importance because it was the first detailed observation which the old cosmological framework simply could not explain away. Something was rotten in the state of astronomy indeed. Tycho’s new star was a small crack in what had been considered a pristine crystalline firmament. There would be others–so many, in fact, that the entire system would soon collapse and shatter. It wasn’t just the heavens which had proven themselves mutable. A revolution was underway, and science, philosophy astronomy–and astrology–would never be the same.

The immutability of the heavens had been discussed and disputed by astronomers throughout Europe with respect to comets (sub– or supralunar?) since Paolo dal Pozzo Toscanelli (1397–1482) viewed them as supralunar based on his observations of the comet of 1456. The observations and reports of the 1572 supernova by many European astronomers only increased an ongoing debate. A debate that was only one part of a general trend to reform astronomy, which started around 1400 and in which everything was up for discussion. The period also saw a revival of Stoic philosophy and cosmology contra Aristotelian philosophy and cosmology. The supernova of 1572 was not the dramatic turning point that Boxer paints it as.

Boxer now delivers, what I regard as the absolute low point of the book, in that he presents the hairbrained theory of Peter Usher that Shakespeare’s Hamlet is “an elaborate astronomical analogy.” He does however backpedal and state, “I enjoy reading this quite a bit, even if I don’t find it very persuasive.” So, why include it at all?

We then move on to a very rapid sketch of the so-called astronomical revolution with the usual Copernicus=>Tycho/Kepler=>Galileo=>Newton cliché. Boxer now allows himself a real humdinger:

            Clearly Tycho’s commitment to a geocentric cosmos ran much deeper than astronomical arguments alone. IN fact, so central was the Earth’s fixity to Tycho’s philosophy that he proposed a compromise cosmology, one in which Mercury, Venus, Mars, Jupiter, and Saturn orbited the Sun, as in the Copernican system, but the Sun and Moon orbited the Earth as in the Ptolemaic system. It sounds ungainly, and Tycho may have been the only person who ever thought otherwise… [my emphasis].

Tycho may have been the only person? A handful of astronomers all independently came up with the so-called Tychonic geo-heliocentric system around the same time, as an alternative to the Copernican system, leading Tycho to accuse others of plagiarism. From about 1620 till about 1660 the majority of European astronomers thought a Tychonic model with diurnal rotation was the most probable system for the known universe.

Boxer finally gets back on course with the next section where he investigates the use of the words, astronomy, astrology, and mathematics to describe either astronomy or astrology as we know them. A very well-done section. This is followed by a section on the Gregorian calendar reform and why it was necessary, relatively good except for a false claim about Copernicus. He writes:

Copernicus cited the prospect of a more accurate calendar as one reason why he hoped (quite wrongly) that his new, Sun-centered theory of the universe might be well received by the Church.

I have no idea where Boxer found this but it’s simply not true. Copernicus’s only connection with the calendar reform was when he was approached around 1520, like many other European astronomers, to contribute to the calendar reform, he declined, stating that one first needed to accurately determine the length of the year. The chapter closes with a brief account of Kepler’s attitude and contributions to astrology, which falsely claims that he rejected astrology at the end of his life. He didn’t, he rejected traditional horoscope astrology most of his life, although he earned money with it, but believed till the end in his own system of celestial influence.

The final section of the book deals with modern forms of astrology. We have the Madame Blavatsky’s Theosophical Society and her creation of spiritual astrology. The creation of the popular twelve-paragraph newspaper horoscope and finally the creation of psychological astrology, first by the theosophist Alan Leo and developed further by psychoanalyst Carl Jung. Here Boxer delivers, what I regard as the biggest error in his entire book. He writes:

Yet the converse opinion–that every good astrologer must also be a good psychoanalyst–is pretty much the default amongst modern astrologers and their clients alike. For the professional astrologer, this represents a tremendous job promotion. A classical astrologer was, first and foremost, a human calculator, one whose most important qualification was his ability to solve long and tedious mathematical equations. [My emphasis]

Here Boxer, the mathematician, shows that he has literally not understood the difference between casting a horoscope and interpreting a horoscope. In fact, in his book he never really addresses the interpretation of horoscopes, which is the real work of a classical astrology. From the few hints that Boxer gives when discussing horary astrology (which he falsely labels judicial astrology) and elective astrology, he appears to think that you just plug in the planetary positions and the horoscopic spits out the interpretation algorithmically. Nothing could be further from the truth. 

Ptolemy writes at the beginning of the Tetrabiblos, I paraphrase, the science of the stars has two aspects, the first deals with the positions of the stars [our astronomy, his Almagest] and is precise, the second deals with their influence [our astrology, his Tetrabiblos], which is not precise. The first involves casting horoscopes and is mathematical, the second with their interpretations and is not mathematical.

If an astrologer, let us say in the sixteenth century the golden age of astrology, casts a full birth horoscope with planetary positions, houses, aspects, lunar nodes (which Boxer doesn’t deal with as being unnecessarily confusing, directions (explained wrongly by Boxer), lots of fortune (which he doesn’t even mention), etc. You have a very complex collection of material that has to be weighed up very carefully against each other. It is highly unlikely that any two professional astrologers would give the same interpretation, each arguing for their interpretation and explaining why the other interpretation is wrong. Very much of this art of interpretation is based on simplel psychology. A court astrologer, who is basically a political advisor, is going to include many psychological, political, and social factors into the interpretation that he delivers up for employer. 

I recently copyedited the translation of a chapter from a thirteenth century Arabic treatise on astrology that dealt with the interaction of the lunar nodes with the houses when practicing elective astrology. The complexity of the interpretive factors that have to be taking into consideration is mindboggling, so please don’t claim that “a classical astrologer was, first and foremost, a human calculator,” it simply isn’t true. 

If you have read this far you might come to the conclusion that my opinion of Boxer’s book is entirely negative, it isn’t. I think there is an excellent, interesting, and important book struggling to get out of a pool of confusion. Boxer’s strength is that of the data scientist and statistician and his sympathetic to astrology statistical analyses of various aspect of astrology are excellent and very much worth reading for anybody interested in the topic. His book cannot be considered a history of western astrology as he simply leaves much too much out. In fact, it is clear that those things he chooses to include are those that give him the possibility to apply his statistical analysis. Is it a “competent overview of astrology”? No, he leaves too much out, for example any competent overview of astrology must include the lunar nodes and their function in astrology and makes too many errors in his presentations of both the history of astrology and astronomy. Most importantly astrology is about the interpretation of horoscopes, a topic that he does his best to avoid.


[1] Alexander Boxer, A Scheme of HeavenThe History of Astrology and The Search for Our Destiny in Data

[2] Although he constantly refers to astrology rather than western astrology, he does state that his book doesn’t deal with other forms of astrology such as Indian or Chinese. 

15 Comments

Filed under Book Reviews, History of Astrology, History of Astronomy

15 responses to “Astrology, data, and statistics

  1. Very interesting, pity about the historical errors.

    However, I’m now left windering what you would recommend on the history of Astrology?

  2. That most valuable of reviews: The one that makes it unnecessary to read a very problematic book. As usual you are very kind, as well as staggeringly authoritative. I would have just thrown it against the wall.

  3. Carmelo

    Great review, I loved reading it.

    I don’t want to pass as a pedant, but in the sequence discussing the etymology of Ptolemy’s Almagest, you typoed “entomology”. (Sorry, it’s too funny not to point it out)

  4. Thank you for your time and honesty with this review. If I may, a few thoughts from the author.

    * First: you have totally misconstrued my take on Manilius where I agree with you 100%. Manilius is merely the first, extended textual description of astrology that survives. He himself can take credit for nothing. I thought that was plain enough in my book. Evidently, I ought to have been more explicit.

    * You object to my description of a stereographic projection as “looking down on a globe from above its North Pole” My statement is a 100% accurate description of a south-pole projection. I made all the diagrams in the book; pretty sure I know what I’m talking about here.

    *As to giving some sections too little, or too much space, that is more a matter of editorial opinion. My main goal was to focus on specific authors, so that I could speak precisely and not generalize. This is certainly true of the chapter on Masha’allah, since we have his text. It was not my aim to cover all of Arabic astrology.

    *You write: “We then get an account [of] Guido Bonatti, who was a Guelph astrologer and author of the Liber Astronomiae, which Boxer tells us, hyperbolically, is the most influential astrology book of the Middle Ages.” 1st, Bonatti was a Ghibelline. 2nd, it’s Thorndike who writes “The most important astrological work produced in Latin in the thirteenth century seems to have been the Liber astronomicus of this Guido Bonatti” (p.826)
    * Similarly for Cecco d’Ascoli, my account follows Thorndike.

    *You write: “Sorry but this statement is garbage the houses are not a coordinate system.” They most certainly are. Not sure how this is controversial.

    *Next you write that I am misreading Regiomontanus. Have you read the Tabulae directionum beyond its title? The introduction and text make clear that the main obstacle to directions was, indeed, domification.

    *”Which system of Houses division is still disputed amongst astrologers and Boxer possesses the impertinence to suggest they should use a particular system because he finds it mathematically the most elegant.” Well, Tycho agrees with me 😉

    * “Copernicus cited the prospect of a more accurate calendar as one reason why he hoped (quite wrongly) that his new, Sun-centered theory of the universe might be well received by the Church.” I have no idea where Boxer found this but it’s simply not true.” Answer: Preface to the De revolutionibus, right at the end. Copernicus couldn’t be more clear.

    * As for astrological interpretation, and interpretation of data in general, this is woven throughout the book. Most explicitly at the very end of Ch. 5 on Ptolemy. But my goal was never to instruct someone how to read a horoscope. Indeed, presumably no 2 astrologers at any time would ever agree on that.

    * Again, thank you for taking the time to read the book. Was really hoping you’d like it. Sad to hear it missed its mark. Be well, ~Alex

    • Ah, the part about Peter Usher got left out. You say this is “the absolute low point of the book, in that he presents the hairbrained theory of Peter Usher that Shakespeare’s Hamlet is “an elaborate astronomical analogy.” Yes, Peter Usher is insane! The meta-theme of the book is how people weave narratives out of data, whether in astrology, data-science, or history. Usher’s theory is a particularly absurd example of this.

    • Boxer:

      “Copernicus cited the prospect of a more accurate calendar as one reason why he hoped (quite wrongly) that his new, Sun-centered theory of the universe might be well received by the Church.” I have no idea where Boxer found this but it’s simply not true.” Answer: Preface to the De revolutionibus, right at the end. Copernicus couldn’t be more clear.

      Ah yes, I had forgotten the Preface to De revolutionibus. Let us examine the evidence:

      In my review I wrote:

      I have no idea where Boxer found this but it’s simply not true. Copernicus’s only connection with the calendar reform was when he was approached around 1520, like many other European astronomers, to contribute to the calendar reform, he declined, stating that one first needed to accurately determine the length of the year.

      Here is what Copernicus writes in his Preface:

      Astronomy is written for astronomers. To them my work too will seem, unless I am mistaken, to make some contribution also to the Church, at the head of which Your Holiness now stands. For not so long ago under Leo X the Lateran Council considered the problem of reforming the ecclesiastical calendar. The issue remained undecided then only because the lengths of the year and month and the motions of the sun and moon were regarded as not yet adequately measured. From that time on, at the suggestion of that most distinguished man, Paul bishop of Fossombrone, who was then in charge of the matter, I have directed my attention to a more precise study of the topics. But what I have accomplished in this regard, I leave to the judgement of Your Holiness in particular and of all the learned astronomers.

      Copernicus is twisting the facts a little. As I stated Paul of Middelburg (Paul bishop of Fossombrone) wrote to quite a large number of European astronomers, including Copernicus, asking them to contribute to a planned calendar reform, Copernicus declined, stating, as already said, that one first needs to determine the length of the year. We know this because the correspondence still exists.

      Here in the Preface, Copernicus states that he has determined new values for the astronomical parameters “lengths of the year and month and the motions of the sun and moon.” He makes absolutely no mention of “his new, Sun-centered theory of the universe,” as these parameters are completely independent of the cosmological system held by the observing astronomer and remain the same irrespective of whether the astronomer is a geocentrist, a heliocentrist, or a geo-heliocentrist, a fact of which Copernicus was well aware.

      In its early reception, astronomers, irrespective of their views on cosmology, did in fact eagerly examine De revolutionibus in the belief/hope that it contained better astronomical parameters than the then most widely used Epitome of the Almagest (Peuerbach/Regiomontanus). Being largely based on the same corrupt data from Ptolemy it didn’t, and this is why Tycho Brahe, and William IV of Hesse-Kassel all set new observation programmes to obtain a completely new set of astronomical observational data.

    • P.S. There is a widespread myth that Copernicus’ length of the year was used for the actual calendar reform, it wasn’t.

    • It was not my aim to cover all of Arabic astrology.

      You actually ignore about 99.9% of Arabic astrology, which means that you book is in no way a simple yet competent overview of astrology.

      • I stand by my choices of what was included / excluded in the book. Here we can agree to disagree. My objection, rather, is that you have been so quick to accuse me of factual inaccuracies — a serious charge. In many of these accusations you yourself have been flat-out wrong. Others are properly arguments over interpretation, all of which I can defend. It’s not my place to argue each of these with you (though I happily would), nor to convince you to like my book. However, I am honor-bound to defend my book from charges of inaccuracy. A zeal for accuracy, detail, and correctness is, I know, a trait that we both share. Very respectfully, ~Alex

    • “* You object to my description of a stereographic projection as “looking down on a globe from above its North Pole” My statement is a 100% accurate description of a south-pole projection. I made all the diagrams in the book; pretty sure I know what I’m talking about here.”

      Yes, but it’s not a description of stereographic projection that would be intuitively obvious to someone who wasn’t already familiar with it. If instead you had written “Imagine a wire-frame model of the Earth with the wires along the lines of latitude and longitude and a point light source at the south pole. Then the shadows cast on a plane surface are a stereographic projection of the sphere” that would be more clear. That you managed to confuse Thony, shows that your description was not as clear as you thought.

      I did a module on crystal optics as part of my physics degree, so I think I understand stereographic projection.

  5. Next you write that I am misreading Regiomontanus. Have you read the Tabulae directionum beyond its title? The introduction and text make clear that the main obstacle to directions was, indeed, domification.

    This does not in any way alter the fact that János Vitéz Archbishop of Esztergom requested a book explaining directions and that this is what Regiomontanus delivered. That he considered a reform of domification to be part of the explanation, doesn't make it the aim of Vitéz's request or of Regiomontanus's fulfilment of that request

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s