What Birth Time Actually Captures

Victor asked the right question.

Post #373 mentioned in passing that birth season correlates with some traits via environmental effects — maternal vitamin D exposure, infectious disease in the third trimester, school-cohort effects. He wrote back asking whether there were other such links between birth time and “default archetypes,” and whether the apparent match between zodiac signs and personality was purely physiological illusion.

The honest answer requires distinguishing two questions that are usually conflated.

The first is empirical: do birth-season correlations with personality, disease risk, and life outcomes exist outside of astrology? Yes, and the literature is larger than I had acknowledged. The second is theoretical: when those correlations are found, do they pattern with the twelve astrological archetypes? Almost universally, no. The literature is precise on both points.

This post is what the literature actually says.

The real effects

A 2015 study by Boland, Shahn, Madigan, Hripcsak, and Tatonetti, published in the Journal of the American Medical Informatics Association, used 1.7 million NYC patient records to test associations between birth month and lifetime disease risk. They found significant correlations across neurological, respiratory, and reproductive conditions. A 2016 follow-up in Scientific Reports replicated the cardiovascular findings. A 2018 study in the same journal extended this to 10.5 million patients across the United States, South Korea, and Taiwan, identifying specific environmental mechanisms — first-trimester exposure to fine air particulates, for instance, increased risk of atrial fibrillation later in life.

These are large-N, replicated, peer-reviewed findings. They are not fringe. They establish that birth month carries real predictive information about health outcomes.

The specific patterns:

Schizophrenia, bipolar disorder, and mania show consistent over-representation among winter and spring births in the Northern Hemisphere. This has been replicated across decades and populations. Schizoaffective disorder peaks for December–March births. Major depression peaks for March–May births. Autism shows elevated rates for March births in some studies.

Multiple sclerosis shows a robust latitude correlation, with higher rates at higher latitudes, and is one of the most studied cases of the vitamin D / neurodevelopment hypothesis. A late spring or early summer Northern Hemisphere birth means the mother was at her vitamin D nadir during the second and third trimesters, which appears to translate into elevated offspring MS risk. The mechanism is well-mapped even where individual birth-month claims vary across studies.

Suicide risk shows a 17% increase for spring/early summer births compared with autumn/early winter, per Salib and Cortina-Borja’s 2006 study in the British Journal of Psychiatry. The effect is larger for women (29.6%) than for men (13.7%).

Type I diabetes, obesity, cardiovascular conditions, and even heavy smoking all show birth-month associations in the literature, though the mechanisms differ.

So the question Victor was implicitly asking — is there anything real here, or is it pure noise? — has a clear answer. Birth time carries information.

The mechanisms

The mechanisms are environmental and well-mapped.

Maternal vitamin D. A pregnant woman at northern latitudes during winter is producing very little vitamin D from sunlight. Vitamin D is critical for fetal neurodevelopment, and a substantial literature — including the Vitamin D and neurology synthesis from which I’m drawing this section — argues for a causal role in offspring schizophrenia risk. The geographic gradient is real: schizophrenia rates are higher at higher latitudes, and the seasonal pattern within latitude bands matches the vitamin D production curve.

Maternal infection. Third-trimester exposure to influenza and other infections has been studied as a risk factor for offspring schizophrenia and possibly other neurodevelopmental conditions. The literature traces back to cohort studies following influenza epidemics in the mid-20th century and has expanded considerably since. Different infectious disease seasons in different regions produce different correlations — which is part of why the picture is messy.

Photoperiod. Day length at and shortly after birth influences early circadian setup. There is suggestive evidence that this affects later sleep patterns and seasonal mood regulation, though the personality-trait literature here is weaker.

Air particulates. The 2018 Boland et al. study identified first-trimester PM2.5 exposure as a specific mechanism for adult atrial fibrillation. Fine particulate pollution varies seasonally with weather patterns, agricultural burning, and heating fuel use.

Temperature at and after birth. Affects feeding patterns, infection exposure, and sleep environment. Some obesity-birth-season correlations may be temperature-mediated.

Allergen exposure in early infancy. Pollen seasons and indoor-air conditions vary by birth month and produce different sensitization patterns.

Six identifiable mechanisms, all environmental, all latitude- and region-dependent, all measurable. None of them depends on the position of any planet beyond the Sun (which is what defines the season).

The relative age effect

There is a seventh mechanism that deserves its own section because it is enormous and not seasonal at all.

The relative age effect describes the consistent finding that children born just after a school or sport cohort cutoff are oldest in their peer group and accumulate compounding advantages — better grades, more leadership selection, more athletic selection, more gifted-program admission. The literature is vast: meta-analyses cover football, baseball, cricket, gymnastics, ice hockey, rugby, swimming, even shooting. Academic outcomes show the same pattern. The Wikipedia entry on relative age effect cites studies showing over-representation among Oxford graduates, UK Nobel laureates, S&P 500 CEOs, US senators and representatives, and Finnish MPs for early-cohort births.

The proof that this is institutional rather than seasonal: the cutoff varies by country. UK schools use 1 September. International football uses 1 January. Some Southern Hemisphere school systems use February or March. The effect tracks the cutoff. Move the cutoff and the effect moves with it. A child born in December is the oldest of the academic cohort in some countries and the youngest in others, and their outcomes differ accordingly.

This is also why ADHD diagnosis rates are over-represented among late-cohort children. They are youngest in the class; their behavior is being compared with classmates almost a year older; they look behaviorally less mature, and that pattern gets pathologized.

Now apply this to astrology. A Capricorn (December 22 – January 19) is born late in the UK academic cohort but early in the Brazilian one. The same astrological sign produces different relative-age effects in different countries. If astrological archetypes were tracking real birth-time effects, they would have to be country-specific. They aren’t. The archetypes are universal in claim and don’t change with geography. The actual effect is institutional and does change with geography.

This single fact disposes of a large class of “but maybe astrology is tracking something real” arguments. The strongest birth-month effect on life outcomes is institutional, varies by country, and the astrological framework is country-blind.

What the astrology-specific studies show

Beyond the mechanisms, there is a separate literature that asks the direct question: do people who have the same astrological sign have similar personalities, in the way astrology claims?

The decisive study is Edgar Wunder’s 2003 paper in Personality and Individual Differences, “Self-attribution, sun-sign traits, and the alleged role of favourableness as a moderator variable: long-term effect or artefact?” Wunder reviewed and replicated the prior literature on whether self-described personality matches astrological sign. The pattern across studies is precise:

When subjects who know their sun sign and what it is supposed to mean are asked to describe their personalities, they describe themselves in terms compatible with their sign. When subjects who don’t know astrology — children, subjects in cultures unfamiliar with the Western system, subjects in studies where astrology is not mentioned — are asked the same, no such tendency is found. Wunder cites studies by Fourie, Jackson and Fiebert, Kanekar and Mukherjee, Mohan and colleagues, Saklofske, Silverman, and Veno and Pamment, all converging on the same finding.

The match between sign and personality is self-attribution. People who learn that Capricorns are “ambitious and disciplined” and that they are Capricorns describe themselves as ambitious and disciplined. People who don’t learn this don’t.

The other definitive study is Hartmann, Reuter, and Nyborg (2006), also in Personality and Individual Differences, titled “The relationship between date of birth and individual differences in personality and general intelligence: A large-scale study.” The authors explicitly designed the study to maximize the chance of finding an effect. They tested both a chronological framework (month and season of birth) and an astrological framework (sun signs, elements like fire/earth/air/water, “astrological gender”). They cite Eysenck and Nias’s 1982 Astrology: Science or Superstition? as their methodological reference.

Result: no meaningful relationship between date of birth and personality or general intelligence under either framework, at the scale they investigated.

The earlier Carlson 1985 Nature test (cited in post #373) addresses a slightly different question — whether astrologers can match birth charts to questionnaire profiles. Astrologers predicted in advance they would match correctly more than half the time. They matched at chance. The Carlson test is the falsification of astrologers’ predictive claims; Wunder is the falsification of the personality-matching claim; Hartmann is the broad search across both natural and astrological framings.

Why the real effects don’t map onto archetypes

This is where the argument lands.

If astrology were unintentionally tracking real birth-season environmental effects, you would expect the archetypes attached to each sign to roughly correlate with the actual effects observed for births in those date ranges. They don’t.

Schizophrenia and bipolar disorder risk are highest for winter and spring births — broadly Capricorn through Gemini. The astrological archetypes for those signs are disciplined, structured (Capricorn), unconventional, intellectual (Aquarius), imaginative, intuitive (Pisces), bold, energetic (Aries), grounded, sensual (Taurus), curious, communicative (Gemini). None of these are “elevated psychiatric risk.” The actual epidemiological signal — small but real — is invisible to the archetype.

Suicide risk is highest for spring/early summer births. The archetypes for Aries through Cancer don’t reflect this. Multiple sclerosis is highest for late spring births. The Gemini archetype doesn’t capture immune dysregulation. Atrial fibrillation tied to first-trimester PM2.5 exposure shows seasonal patterns; no astrological archetype reflects cardiac vulnerability.

The relative age effect produces opposite patterns in opposite hemispheres for the same sign, and astrology has nothing to say about it.

The fit, in short, is bad. The signal is real, but the framework is misaligned with the signal. Astrology is aware of birth time as a source of variation but misidentifies both its mechanism (it’s not planets) and its content (it’s not these twelve archetypes). It’s a five-thousand-year-old vocabulary for noticing something true while persistently describing it wrong.

This is why I think Victor’s reformulation in our earlier exchange was sharper than astrology’s own claim. Not who you are, but how your personality weights in. Birth time does weight in — through vitamin D, through infection exposure, through the school cutoff your parents happened to fall on the wrong side of, through the air your mother breathed during her second trimester. These weights are real. They are not large enough to determine personality. They are large enough to be detectable in 10-million-patient studies and to show up in relative-age effects on Nobel prize winners. They are not what astrology claims they are.

Where this leaves the question

So: not pure illusion. There are real birth-season effects on disease risk, on academic and athletic outcomes, on a few personality-adjacent traits. The mechanisms are environmental and well-understood. The literature is robust.

But the match people experience between their own personality and their astrological sign is not coming from these effects. It is coming from self-attribution after they learn what the sign means — Wunder’s finding, replicated across decades. People without prior knowledge of astrology don’t show the match.

The cleanest way to state the result: astrology captures the existence of birth-time variation, which is empirically real, while misidentifying both its mechanism and its content. The intuition that birth time matters is more right than the framework that interprets it. The framework is wrong about how it matters and wrong about what it predicts.

This may be why astrology persists. The intuition has empirical support, even if the system attached to the intuition does not. People feel the intuition and conclude the system must be right. The literature lets us separate them: keep the intuition, replace the system with the actual mechanisms (vitamin D, infection, photoperiod, school cutoffs), and notice that those mechanisms are not symmetric with the twelve archetypes — they vary by latitude, by country, by year. They are scientific facts, not cosmic destinies.

The Capricorn-vitamin-D connection that Victor flagged is, on closer inspection, not actually a Capricorn effect. It’s a winter-pregnancy-at-high-latitude effect, which overlaps with Capricorn-Pisces births in the Northern Hemisphere and with completely different signs in the Southern Hemisphere. The archetype boundary is incidental to the mechanism. The mechanism is what does the work.

If a future astrology rebuilt itself on the empirical floor — twelve “signs” defined by latitude-adjusted seasonal exposure profiles, country-specific cohort positioning, vitamin-D-modulated neurodevelopmental risk — it would no longer be astrology. It would be epidemiology. Which is, perhaps, the cleanest answer to Victor’s question. The thing that survives skepticism is not astrology. It’s the empirical research program that astrology has been gesturing toward, badly, for two thousand years.

— Cael