A naturalistic argument for birth timing, prenatal development, and the strange persistence of human archetypes
I have heard the standard arguments against astrology for most of my life. People recognize themselves in vague descriptions. They remember the hits and forget the misses. Positive traits flatter the reader. Expectations influence interpretation. Once someone knows his sign, he can slowly absorb its description into his identity. All of that is plausible, and some of it is certainly happening.
There is still something left over for me.
I have known enough people closely, across years and circumstances, to keep seeing the same basic archetypes appear. They are far more interesting when their flattering language is removed. Each sign carries advantages that can mature well and liabilities that can develop badly. Independence can harden into senseless contrarianism; emotional sensitivity can deepen perception or destabilize judgment, while persistence can mature into exceptional consistency or stubbornness. When I look at people this way, the negative expression adds explanatory value because it points to individual development along a spectrum within the same archetype. I will return to this shortly.
My own case brought the issue into focus. I am an Aquarius. A description of Aquarian intelligence emphasized unconventional reasoning, intellectual independence, rapid conceptual movement and the ability to apprehend whole systems. I already knew the usual psychological explanations for why horoscope descriptions feel accurate. Yet that particular profile struck too close to traits I had identified in myself long before caring much about astrology. More importantly, I had already seen similarly recognizable patterns in other people.
Personal observation proves very little by itself. Large studies have failed to find meaningful relationships between zodiac sign or date of birth and psychometric measures of personality and intelligence. A large-scale analysis combining more than 15,000 participants found no relevant association, while an earlier representative sample of 3,074 young men likewise found no correlation between zodiac signs and measured personality.[1,2]
I take those results seriously. I also think they leave room for a different question.
Suppose, as a mental exercise, that the recurring archetypes I think I see eventually survive better investigation. Suppose the effect is weaker and more statistical than popular astrology suggests, yet real enough that birth timing carries some information about temperament. What would follow?
The first mistake would be assuming that the traditional explanation must survive with the observation.
A civilization can observe something correctly and explain it badly.
That possibility changes the entire problem.
The Observation Before the Explanation
When an old belief contains an implausible mechanism, modern thinking often disposes of the entire package at once. Sometimes that is deserved. Yet an observation and the explanation attached to it are different kinds of claims.
People knew that certain plants relieved pain long before biochemistry could explain why. Sailors could recognize relationships among winds, clouds and coming weather without understanding atmospheric physics. Breeders altered plants and animals for thousands of years before the mechanisms of inheritance were understood. Human beings are capable of accumulating practical correlations while constructing completely wrong stories about their causes.
Astrology could belong to that category.
The constellations themselves might be irrelevant. The mythological meanings assigned to planets might be later interpretation. Even many claims made by professional astrology could fail while some older observation survived underneath them: people born during different portions of the year may show slightly different distributions of temperament.
If that were true, calling the phenomenon “astrology” would already prejudice the investigation. It would make us search among the symbols rather than asking what physical variable the symbols accidentally index.
That brought me to a simpler thought.
A date is a position in time. Time on Earth is inseparable from motion.
Every date locates Earth at a particular point in its annual orbit around the Sun.
Birth May Be the Wrong Event
At first it seems obvious to ask what could happen on the day someone is born. That may be the wrong question.
A February birth also tells us approximately when conception occurred. It tells us where early embryonic development fell within the annual cycle, where major periods of fetal brain development occurred, which months contained late gestation, and when the transition into independent life happened.
Move the birth date forward by eight weeks and the entire prenatal sequence moves with it.
So the relevant variable might never have been the date of birth itself.
Birth could simply be the easiest visible marker for a biological history extending backward through gestation.
That distinction matters because prenatal development is highly sensitive to timing. Maternal stress during pregnancy has documented associations with later neurodevelopment, meaning the development of the brain and nervous system, and contemporary reviews emphasize that the consequences vary according to the gestational period during which exposure occurs. Different stages carry different vulnerabilities because the fetus is changing rapidly throughout pregnancy.[3]
This gives us a general principle with much wider implications.
A weak environmental influence can matter differently depending on what the organism is doing when it encounters that influence.
The fetus during week eight is biologically different from one during week thirty. An external variable could remain continuous while its developmental consequence changes because the receiving organism has changed.
This is exactly what makes the birth-timing hypothesis more interesting than a crude monthly effect.
I am proposing a sequence.
Sun–Earth phase → changing geophysical conditions → maternal biological response → fetal developmental window → persistent bias in temperament
Here, geophysical conditions simply means physical conditions associated with Earth and its surrounding environment, particularly magnetic and solar-related changes.
Birth date would serve as the timestamp from which much of that prenatal sequence could be reconstructed.
The question then shifts again. What environmental condition is sufficiently global, recurrent and physically real to remain plausible if the same broad archetypes appear across different climates and hemispheres?
The Sun is the obvious place to start.
The Sun Without Astrology
February in Canada and February in Australia have very different weather. Their local day length, temperature and ecology differ as well. Yet both locations occupy the same Earth at the same point in its orbit.
That immediately separates two classes of explanation.
A climate-based effect should change substantially with geography. A phenomenon tied to Earth’s annual relationship with the Sun could remain global.
Earth follows an elliptical orbit. Its closest point to the Sun, called perihelion, is roughly 147.09 million kilometres away. At aphelion, its farthest point, that distance increases to roughly 152.10 million kilometres. Earth therefore returns each year through a repeatable orbital cycle whose physical properties vary continuously.[4]
Distance is probably too simple to explain much by itself. More interesting is the interaction between the Sun and Earth’s magnetic environment.
The Sun continuously emits electromagnetic radiation—energy carried by electromagnetic waves, including visible and ultraviolet radiation—and solar wind, a stream of electrically charged particles flowing outward from the Sun. The solar wind also carries the interplanetary magnetic field, the extension of the Sun’s magnetic field outward through the solar system. Solar activity rises and falls through an approximately eleven-year cycle as well.[5]
Earth interacts with this changing solar environment through its magnetosphere, the region surrounding the planet in which Earth’s magnetic field dominates. The geometry of that interaction varies through the year. The Russell–McPherron effect, first described in 1973, identifies one important mechanism behind the strong semiannual pattern in geomagnetic activity—disturbances and variations in Earth’s magnetic field. As Earth travels around the Sun, the orientation between the magnetic field carried by the solar wind and Earth’s own magnetic field changes in a predictable way.[6]
None of this establishes astrology. It establishes something narrower and more useful: Earth’s physical environment contains recurring annual and multi-year variation arising from solar-terrestrial interaction.
The next problem is biological.
An electromagnetic or geomagnetic variation matters to this hypothesis only if biology can detect it and translate it into a physiological difference.
Here the evidence grows intriguing while remaining preliminary.
A small study of twenty-five healthy people at 70° north reported an association between geomagnetic activity and melatonin secretion. Another investigation involving 153 men found higher geomagnetic activity associated with lower overnight excretion of a melatonin metabolite—a breakdown product of melatonin that can be measured in urine. These findings fall far short of proving a pathway into temperament, yet they demonstrate that the proposal itself is biologically testable: human neuroendocrine physiology—the interaction between nervous-system signaling and hormonal regulation—may respond, under some conditions, to changes in the geomagnetic environment.[7,8]
Melatonin then offers a particularly interesting bridge into pregnancy.
Human maternal melatonin crosses the placenta readily and enters fetal circulation. Research on pregnancy physiology describes maternal melatonin as one source of temporal information available to the developing fetus, with maternal circadian rhythms—the roughly twenty-four-hour biological cycles governing sleep, hormones and many other bodily processes—helping organize fetal biological timing.[9]
Now the speculative chain has recognizable parts.
The Sun changes, Sun–Earth conditions respond, and some portion of those changes may reach maternal physiology. Maternal signals then reach the fetus while it passes through different developmental stages.
Every component of that chain belongs to ordinary physics and biology, even where the causal connections between them remain uncertain.
The missing link is enormous: evidence that these variations alter human neurodevelopment in a persistent way that later corresponds to zodiac archetypes. Existing science has yet to establish that.
But the hypothesis no longer requires magic.
It requires an unidentified or poorly characterized developmental effect.
That is a radically different claim.
Interestingly, even the general idea has a minor precedent. A 1975 paper discussing an unexplained birth-date/personality finding explicitly considered climatic, dietary and cyclical geomagnetic events as possible influences on prenatal or neonatal development. The particular personality result never grew into established science, yet the causal possibility itself has been considered before.[10]
Why Would There Be Twelve?
This is where the model initially seems to fail.
Nothing obvious in solar physics gives us twelve distinct human states per year. Earth completes one orbit. The major geomagnetic pattern discussed above has a strong semiannual component. The Sun has its own roughly eleven-year activity cycle. The Moon follows yet another rhythm.
Where would Aries, Taurus, Gemini and the rest come from?
I think the expectation of twelve literal physical states is the mistake.
There may be no twelve states in nature.
Imagine instead a continuous annual curve. Earth’s geophysical environment changes gradually. At the same time, the fetus moves through a sequence of developmental stages, each with its own sensitivity. The relevant output is therefore created by the interaction of two changing systems.
A signal encountered early in pregnancy could have one consequence; the same signal months later could have another. Additional environmental variables could amplify, reduce or alter the outcome depending on which stage of development they encounter.
The resulting annual pattern could grow complicated very quickly even if every underlying input remained continuous.
Ancient observers would have had no access to magnetometers, endocrinology, fetal neuroimaging or statistical models. They had people, dates, generations of memory and a sky divided into recurring periods.
Twelve categories may simply have been a human compression of a continuous phenomenon.
We do this constantly. Temperature varies continuously, yet we speak of cold, mild and hot conditions. Color varies along continuous physical dimensions, yet language divides it into named regions. Personality itself is commonly described through categories even when underlying traits exist along gradients.
The zodiac could work the same way.
Aquarius, under this interpretation, would have no independent physical existence. It would designate a region along an annual developmental continuum in which certain temperamental biases occurred somewhat more often.
This would also predict fuzzy borders. Someone born one minute after a zodiac boundary should never undergo an abrupt biological conversion. If an effect exists, transition across adjacent dates should be gradual.
That is exactly the kind of prediction a natural model ought to make.
Why Two People With the Same Sign Can Be So Different
The next objection turns into one of the model’s strengths.
If an Aquarian disposition exists, why are Aquarians so different from one another?
Because an annual solar phase would represent only one input among many.
The Sun itself varies from year to year. Its approximately eleven-year cycle changes the frequency and intensity of solar activity, meaning two pregnancies occupying similar calendar months in different years can occur under different heliophysical conditions—the changing physical conditions associated with the Sun and its interaction with the surrounding space environment.[5]
The Moon supplies another moving variable. Its phase drifts through the solar calendar, which makes it unsuitable as the primary source of a stable February-type pattern yet potentially relevant as a secondary modifier. Human evidence for lunar effects remains inconsistent. A controlled laboratory study reported changes in sleep and melatonin around the full moon, while a much larger population study of 2,125 people found no significant effect of lunar phase on subjective or objective sleep measures.[11,12]
Then biology introduces far greater variation than either.
Two fetuses can encounter similar external conditions and respond differently because their genomes differ. Their mothers differ. Placental function differs. Nutrition, illness, stress, medication and countless contingent events alter the developmental environment. Prenatal stress alone offers a known example of how maternal experience can reach developmental outcomes through timing-dependent biological pathways.[3]
After birth, the divergence expands dramatically.
Family, culture, education, relationships, trauma, opportunity and personal choice act upon whatever predispositions existed beforehand.
This is why I find the traditional distinction between a sign’s “higher” and “lower” expression more psychologically plausible than it first appears.
A temperament carries possibilities rather than moral instructions.
Intellectual independence can mature into original thought or deteriorate into opposition for its own sake. Emotional intensity can support unusual psychological perception while also increasing susceptibility to reactive judgment. The same basic tendency can produce admirable and destructive forms depending on everything that acts upon it afterward.
That gives the archetype a limited role. It describes a bias in the starting conditions rather than a finished person.
I find that more plausible than either extreme: a birth date dictating character, or prenatal timing contributing absolutely nothing to later temperament.
A Better Way to Think About Astrology
Under this model, conventional astrology may have confused several layers of reality.
Ancient observers encounter recurring human types. Those types correlate imperfectly with portions of the annual calendar. Calendar periods are associated with constellations and mythological symbols. Later generations inherit the symbols together with causal stories about planets, signs and personal destiny.
Eventually the explanation grows far beyond the original observation.
That historical progression would be completely human.
Once a useful classification acquires religious, symbolic or predictive meaning, people rarely preserve a clean division between what was observed and what was inferred from it.
This also gives us a way to separate the stars from the timing.
Earth’s axis undergoes precession, a slow wobble in the direction of its rotational axis that takes roughly 26,000 years to complete. Because of that motion, the apparent stellar backdrop gradually shifts relative to Earth’s seasonal and orbital reference points.[4]
If a genuine birth-timing effect arises from Sun–Earth physics, the effect should follow the relevant physical phase of Earth’s annual cycle. The distant constellation would drift in symbolic importance while the local physics continued.
That is a powerful distinction.
The stars could have served as the ancient clock face.
A clock face tells time without causing time.
Something similar may have happened with astrology. Human observers could have used visible astronomical positions to label recurring periods while misunderstanding why those periods correlated with anything biological.
This possibility gives astrology a naturalistic interpretation without requiring the symbolism itself to exert causal power.
Numerology and the Problem of Hidden Proxies
Once I began thinking this way, other systems usually grouped under pseudoscience looked different as well.
They should never all receive the same rescue attempt. Some have plausible routes to genuine information. Others have much weaker ones.
Numerology offers a useful example.
Birth-date numerology begins with a real variable: the time at which someone was born.
If birth timing contains biological information, then any numerical transformation of the date starts with a variable that already contains some signal. The later arithmetic could be causally meaningless while retaining fragments of information from its input.
This gives us an immediate test.
Write the same birth in the Gregorian calendar, the Hebrew calendar, the Islamic calendar or another date system. The digits change. The physical event stays exactly where it occurred in Earth’s history.
A genuine biological effect should follow the invariant physical event—the same occurrence regardless of how a calendar represents it.
So if a numerological pattern survives only under one arbitrary calendar notation, the digits themselves deserve little causal importance. If some birth-date numerology nevertheless predicted traits above chance, I would first suspect that its arithmetic had accidentally grouped portions of meaningful temporal cycles.
The numbers might be a lossy encoding of time: a representation that preserves some temporal information while discarding much of the original detail.
Name numerology faces an even larger problem. A person’s name can be transliterated—rewritten from one writing system into another—and produce different numerical values. Any effect there would require another explanation. Parental naming choices contain information about family, culture, religion, generation and social environment. A numerical system might accidentally classify some of those patterns while assigning mystical importance to the calculation.
Again, the same principle applies: separate the information contained in the input from the story attached to the transformation.
Palmistry belongs to a more interesting category.
A hand is a biological phenotype—the observable physical result of genes interacting with development and environment. Its morphology develops under genetic and prenatal influences. If some developmental variables influence both neural traits and physical characteristics of the hand, correlations are possible in principle. In such a case, the hand would function as a record of development.
That gives palmistry a conceivable natural pathway without granting claims that a crease predicts the year of a marriage or the timing of death.
The mistake would be moving from “this physical trait correlates with another trait” to “this mark contains a coded biography of the future.”
Tarot and similar divination systems are different again.
I suspect their strongest potential lies less in hidden information about external events and more in the psychology of interpretation. Give a person an ambiguous image and ask what it means. His concerns determine what attracts attention. Memory supplies associations. Fear and desire alter interpretation. The symbol can therefore provoke genuine self-discovery even if the card itself knows nothing.
That would make Tarot closer to an instrument for eliciting projection than a device receiving information from the future.
The interesting point is that several practices dismissed under one word—pseudoscience—could contain entirely different mixtures of observation, proxy information, psychological projection, cultural inheritance and fantasy.
Treating them all identically may be as intellectually lazy as believing all of them.
What This Hypothesis Would Actually Predict
A serious hypothesis has to risk failure.
The model I am proposing makes several predictions that conventional horoscope reading rarely considers.
Gestational age should matter. If birth date mainly indexes a prenatal sequence, a premature baby and a full-term baby born on the same date have experienced different developmental histories. Conception timing and developmental stage should eventually predict any real effect better than the birthday alone.
Exact sign boundaries should also weaken under close measurement. Biology usually changes through gradients. If personality statistics jumped abruptly when the calendar crossed from one zodiac sign into another, my continuous developmental model would struggle to explain it.
Year of birth should carry additional information because the Sun’s activity changes across its magnetic cycle. Lunar state might account for a smaller layer of within-period variation if any reproducible human pathway exists. Local environmental variables should add further deviations where climate, diet, infection or day-night exposure matters.
The most revealing test may concern geography. If the primary association follows ordinary seasons, the pattern should change substantially across hemispheres. If the broad pattern remains tied to the same annual Earth–Sun phase across latitude and hemisphere, a global geophysical explanation gains plausibility.
And because axial precession shifts the apparent stellar background over very long periods, an effect tied to Earth’s annual physics should ultimately follow the physical cycle more faithfully than the distant constellations used to name it.[4]
These predictions could all fail.
That possibility is exactly why I find the model worth considering.
It converts an argument about belief into an argument about mechanism.
The Part I Find Hardest to Dismiss
I have no need for astrology to be literally true in the form it has been handed down.
In some ways I find the alternative more interesting.
Human beings may have spent thousands of years living inside recurring natural cycles, observing one another closely, preserving patterns through myth and classification, while lacking the instruments required to identify the physical variables underneath them.
They could easily have confused the marker with the cause.
Perhaps the zodiac never described twelve invisible powers acting upon souls. Perhaps it was an early human attempt to classify statistical differences associated with when development occurred inside a continuously changing planetary environment.
The Sun would be the first candidate because it dominates Earth’s external energy environment and drives a vast range of terrestrial cycles. Its magnetic interaction with Earth adds annual and multi-year variation that could, in principle, be mediated through maternal physiology. Prenatal development supplies the timing sensitivity required for the same recurring input to have different consequences at different stages, while lunar state, genetics and maternal condition could modify the result. Life after birth would determine what any initial disposition eventually turns into.
None of that establishes the zodiac.
It does something I find more valuable: it shows how a phenomenon that initially sounds absurd can be reformulated into a physical question with identifiable causal gaps and testable consequences.
There is a broader lesson here about old knowledge.
Modern skepticism is strongest when it separates observation from explanation. It weakens when the failure of one explanation is treated as proof that every observation surrounding it was imaginary.
Ancient people were capable of being wrong about causes while remaining perceptive about patterns.
We are capable of the opposite error: understanding enough science to reject their explanations, then assuming we have therefore explained away everything they saw.
If the archetypes eventually collapse under better investigation, that settles the issue.
But if some residue survives—if birth timing genuinely predicts aspects of temperament after the ordinary psychological explanations have been stripped away—then I would look first toward the developmental history before birth and the changing physical environment through which that development occurred.
The zodiac may turn out to have been looking in roughly the right direction while telling the wrong story about what it saw.
References
[1] Hartmann, P., Reuter, M., & Nyborg, H. (2006). “The relationship between date of birth and individual differences in personality and general intelligence: A large-scale study.” Personality and Individual Differences, 40(7), 1349–1362. DOI: 10.1016/j.paid.2005.11.017.
[2] Angst, J., & Scheidegger, P. (1976). “Signs of the zodiac and personality.” Sozial- und Präventivmedizin, 21(1), 39–41. DOI: 10.1007/BF01994375.
[3] Collins, J. M., et al. (2024). “Prenatal stress impacts foetal neurodevelopment: Temporal windows of gestational vulnerability.” Neuroscience & Biobehavioral Reviews, 164, 105793. DOI: 10.1016/j.neubiorev.2024.105793.
[4] NASA Science. “Chapter 2: Reference Systems.” Basics of Space Flight. Used here for Earth’s perihelion, aphelion and approximately 26,000-year axial precession.
[5] NASA Science. “Sun: Facts.” Used here for the solar wind, the Sun’s interplanetary magnetic field and its approximately eleven-year activity cycle.
[6] Russell, C. T., & McPherron, R. L. (1973). “Semiannual variation of geomagnetic activity.” Journal of Geophysical Research, 78(1), 92–108. DOI: 10.1029/JA078i001p00092.
[7] Weydahl, A., Sothern, R. B., Cornélissen, G., & Wetterberg, L. (2001). “Geomagnetic activity influences the melatonin secretion at latitude 70 degrees N.” Biomedicine & Pharmacotherapy, 55 Suppl. 1, 57s–62s. DOI: 10.1016/S0753-3322(01)90006-X.
[8] Burch, J. B., Reif, J. S., & Yost, M. G. (2008). “Geomagnetic activity and human melatonin metabolite excretion.” Neuroscience Letters, 438(1), 76–79. DOI: 10.1016/j.neulet.2008.04.031.
[9] Tamura, H., et al. (2008). “Melatonin and pregnancy in the human.” Reproductive Toxicology, 25(3), 291–303. DOI: 10.1016/j.reprotox.2008.03.005.
[10] Pellegrini, R. J. (1975). “Birthdate psychology: a new look at some old data.” The Journal of Psychology, 89(2), 261–265. DOI: 10.1080/00223980.1975.9915760.
[11] Cajochen, C., Altanay-Ekici, S., Münch, M., Frey, S., Knoblauch, V., & Wirz-Justice, A. (2013). “Evidence that the lunar cycle influences human sleep.” Current Biology, 23(15), 1485–1488. DOI: 10.1016/j.cub.2013.06.029.
[12] Haba-Rubio, J., et al. (2015). “Bad sleep? Don’t blame the moon! A population-based study.” Sleep Medicine, 16(11), 1321–1326. DOI: 10.1016/j.sleep.2015.08.002. The study included 2,125 participants and found no significant lunar-phase effect on sleep.