How we write horoscopes

The sun-sign column was born in a single Sunday newspaper feature: R. H. Naylor, Sunday Express, 24 August 1930, "What the Stars Foretell for the New Princess." The piece was a chart for the infant Princess Margaret, commissioned by editor John Gordon as a one-off curiosity. The reader response was so large that the paper kept Naylor on, and within months he was producing a weekly column keyed to the twelve signs — the format the entire modern genre still uses. Nicholas Campion documents the spread in A History of Western Astrology vol II (Bloomsbury, 2009): by 1933 Paul Clancy had launched American Astrology magazine in New York, and by the late 1930s the sun-sign column was a fixture on both sides of the Atlantic. Patrick Curry's A Confusion of Prophets: Victorian and Edwardian Astrology (1992) traces the longer popular-press lineage that made Naylor's column landable. The horoscope as we read it on a phone today is not a survival from antiquity. It is a Depression-era journalistic invention, ninety-five years old, designed to fit one column-inch per sign.

A sun-sign column is not a horoscope in the older technical sense — it is a wire-service condensation of one. In pre-modern practice, a horoscope was a full chart cast for a specific moment: ascendant, twelve houses, every planet placed by degree, aspects calculated. Campion vol II spells out the gap. Hellenistic astrologers read whole charts for individual clients; medieval and early-modern astrologers read elections, nativities and mundane charts. None of them read a paragraph keyed only to which thirty-degree slice of the ecliptic the Sun occupied at your birth. That format did not exist until Naylor needed something a newsroom could print every week. The twelve-paragraph daily is a publishing constraint dressed as a tradition. It uses real astrological vocabulary — signs, houses, transits — and reduces it to a horoscope-shaped paragraph that fits between a crossword and a cartoon. Knowing this changes how to read one: as a piece of editorial writing built on a thin sky-anchor, not as a personalised chart reading.

Geoffrey Cornelius reframed the question in The Moment of Astrology: Origins in Divination (Penguin Arkana 1994; 2nd ed Wessex Astrologer 2003): astrology is a divinatory practice, not a causal one. A prediction is a causal claim — because Mars enters Aries on Tuesday, therefore you will quarrel. A forecast in the divinatory register names a pattern in the sky and asks what it might illuminate for the reader who is paying attention to it. Cornelius does not claim this validates astrology in a scientific sense. He reframes what astrology was ever actually doing: reading a moment, not predicting a billiard-ball outcome. The honest editorial alternative to "you will…" is "this may manifest as…" — a phrase that names the transit, names the field of meaning it traditionally signifies, and hands the reader the work of recognising it in their own life. We use that register deliberately. Every forecast on this site is built on "this may manifest as," because that is the only sentence frame compatible with what astrology, on its strongest internal account, actually does.

The strongest empirical tests of astrology — as a causal claim — return a clean negative result, and we say so plainly. Shawn Carlson, "A double-blind test of astrology," Nature 318:419-425 (5 December 1985), ran the methodologically tightest test on record: thirty experienced astrologers, blind-matched to natal charts and California Psychological Inventory profiles, performed at chance. Geoffrey Dean and Ivan W. Kelly, "Is astrology relevant to consciousness and psi?," Journal of Consciousness Studies 10:6-7 (2003), 175-198, reviewed forty years of subsequent tests across more than 2,000 subjects and reached the same finding: no astrologer outperforms chance under controlled conditions. The longer technical treatment lives at /astrology/is-astrology-real. We refuse to dodge this. Astrology read as physical causation — planets exerting forces, signs determining personality — fails every clean test it has been given. What survives, on Cornelius's account, is something else: a symbolic practice for reading a moment. That is the practice we write from. Naming the negative result is part of writing horoscopes honestly.

Our editorial constitution has four pillars, each tied to a technique you can audit. First, every daily and weekly horoscope is transit-anchored — keyed to the actual transit dominating the sky window we are writing about, with the technique explained at /astrology/transits. No transit, no horoscope. Second, our forecasts are written to be read by rising sign as well as sun sign, because rising-sign keying produces a forecast that matches the houses being activated in a chart rather than only the solar overlay; the longer argument lives at /astrology/rising-sign. Third, every new moon and full moon gets a lunation report — phase, sign, the houses it lights up by ascendant — anchored to /astrology/lunation. Fourth, we publish dated eclipse and ingress forecasts: the four annual ingresses and the eclipse-season windows, with the mechanics at /astrology/eclipses and /astrology/ingresses. All four are written in the "this may manifest as" register from §3. That is what we ship. None of it requires you to believe in causation; it requires you to read the sky as a symbolic field and your life as something happening underneath it.

Four things we refuse, even though they would be cheaper or more clickable than what we publish. One: AI-generated personalised dailies keyed to birth time. The personalisation theatre — "your unique forecast based on your exact moment of birth" — runs into the Carlson and Dean & Kelly evidence head-on and uses a register Cornelius spent a book repudiating. Two: prescriptive "you will" claims. "You will receive money on Wednesday" is a causal prediction, fails every controlled test, and is the register honest astrology has been refusing for thirty years. Three: pure sun-sign mood reads with no sky anchor — twelve paragraphs of feelings-coded prose where any line could be swapped between signs without changing meaning. If there is no transit, lunation or ingress driving the read, we do not write the read. Four: "for entertainment purposes only." That disclaimer is the genre's central act of bad faith — pretending the page is a joke so no one has to defend it. We would rather defend what we publish than hide behind a footer cliché.