What is Jyotisha?

In the year 499 of the Common Era, in Kusumapura near what is now Patna, a man of twenty-three finished a book of a hundred and twenty-one verses and, in one of them, set the earth spinning.

His name was Aryabhata, and he gives his age himself: three thousand six hundred years of the Kali era had elapsed, he writes, and he was twenty-three. The verse sits in the Golapada, the section on the sphere. In the reading of K. S. Shukla and K. V. Sarma it says this. A man in a boat, moving forward, sees the unmoving objects on the bank sliding backwards. In just that way, a man at the equator sees the fixed stars moving steadily westward. The great wheeling of the heavens is an illusion produced by our own motion. The stars are still. We are the ones turning.

He was not guessing. In the same book he gives the earth’s rotations in a mahayuga, a great age: 1,582,237,500. The earth still sits at the centre of his universe. It simply turns. Divide it out and the sidereal day implied is twenty-three hours, fifty-six minutes and 4.1 seconds. The modern value is 4.09 seconds. He explains eclipses as shadows, and in the mechanism he never once mentions Rahu, the demon the tradition held responsible.

A hundred and twenty-nine years later, in 628, Brahmagupta wrote the Brahmasphutasiddhanta, whose eleventh chapter is a chapter-length demolition of Aryabhata. Brahmagupta rejects the rotating earth. He keeps the earth fixed. He needles Aryabhata for computing the lunar nodes as an eighth graha while denying Rahu, which is a fair hit.

And the tradition sided with Brahmagupta.

Be fair to him. Much of that chapter is real technical criticism, and it was Brahmagupta’s book, carried to Baghdad, that gave the world its numerals. But the fact stands. The earth stopped turning in Indian astronomy for the better part of a thousand years, and being right had nothing to do with it.

Everything difficult about jyotisha is already in that story. The astronomy is real, and it is magnificent. The tradition is not one voice. And the thing most people mean today by “Vedic astrology” is not the thing Aryabhata was doing.

The word, and the two things it names

Jyotisha means the science of the lights, from jyotis, a luminary. It is one of the six Vedangas, the limbs of the Veda: phonetics, metre, grammar, etymology, ritual, and jyotisha. The old body-metaphor assigns jyotisha the eyes. It is the limb that sees when.

That is not decoration. Vedic religion is a religion of time-bound acts: the new-moon and full-moon sacrifices, the seasonal rites, the year-long sattras with their solstitial turn. A sacrifice performed on the wrong day is not the sacrifice. Someone had to keep the count, and that someone invented Indian astronomy.

Now set beside it the other thing the word names: a kundali, a birth-chart, cast this morning for a child in a Pune hospital, with twelve signs and twelve houses and nine grahas and a dasha sequence running out to a hundred and twenty years.

Both are called jyotisha. Roughly a thousand years separate them, and one of them came through Alexandria. Collapsing the two is the commonest error made about this material.

Lagadha’s calendar

The oldest surviving Indian astronomical text is a short, terse thing attributed to a sage named Lagadha, in two recensions of about thirty-six and forty-three verses. That is all of it. The standard edition is T. S. Kuppanna Sastry’s, edited by K. V. Sarma (1984).

It is not modest. As the crests stand on the heads of peacocks, it declares, as the gems stand on the hoods of serpents, so does jyotisha stand at the head of the auxiliary sciences.

The boast is about a calendar, and the calendar is real engineering. Twelve lunar months run about eleven days short of the solar year, so a festival fixed to the moon walks backwards through the seasons until it lands in the wrong one. Lagadha’s answer is a wheel of five years, which he calls a yuga. Beware the word. This yuga is a five-year calendrical unit, not the cosmic mahayuga of the Puranas that runs to hundreds of thousands of years, which we treat in the chapter on the wheel of ages.

Inside the wheel the text counts 1,830 civil days, 62 synodic months from new moon to new moon, 1,860 tithis or lunar days, and two intercalary months, adhikamasa, slotted in to drag the moon back into step with the sun. It does not quite work: 62 synodic months come to about 1,830.9 days, so the scheme loses nearly a day a cycle. That matters, because it is what makes the later mathematical astronomy legible as an achievement rather than an inheritance.

And here is the load-bearing fact of this article. The Vedanga Jyotisha contains no zodiac. No houses. No birth-charts. No fate read from the planets. The five visible planets barely appear. It is about when to sacrifice, and nothing else. The fuller treatment lives in our chapter on the Vedanga of meaning and the measure of time.

How old is it? Two questions, kept apart. The text describes a winter solstice at a point the sky no longer holds, and back-computing it gives a date consistent with roughly 1400 BCE, some calculations landing nearer 1180 BCE. When the surviving text was fixed is contested: Kuppanna Sastry and Rajesh Kochhar read the astronomy as dating the document, Pingree placed the extant recension around 400 to 300 BCE, holding that the old solstice’s bearing on the text’s date “is not evident,” and Michael Witzel puts it in the last centuries BCE on linguistic grounds. The data are old. The document is probably younger. Whether the one dates the other is unsettled.

One recent Indian result deserves naming. Mandadi, Sooryanarayan and Ramasubramanian, at IIT Bombay, recomputed the text’s own daylight-length function in 2025 and found it fits a latitude of about 29 degrees north, roughly Kurukshetra, and fails at Mesopotamian latitudes. That will matter shortly.

The eyes of the moon

Before the zodiac there were the nakshatras, and they are the older, more Indian sky. A nakshatra is a lunar mansion, one of twenty-seven segments of the moon’s monthly path, each named for a star or asterism. Twenty-seven, because the moon takes a little over twenty-seven days to come back round. Some lists give twenty-eight, adding Abhijit, which others drop. The earliest full lists are in the Taittiriya Samhita (4.4.10.1-3) and the Atharvaveda (19.7).

The Rigveda gives no list. It gives one line. In the great wedding hymn, Rigveda 10.85.13: aghasu hanyante gavo, ‘rjunyoh paryuhyate. The oxen are slain in Magha. The bride is led home in Arjuni.

The wedding has a star. That is the oldest surviving link in India between a position in the sky and the timing of a human ceremony, and the direct ancestor of every muhurta consulted for a wedding in India this week.

One caution, loudly. Star positions shift against the seasons as the earth’s axis wobbles, and the shift can be run backwards to date a text. It is a legitimate tool and a notorious engine of fantasy. The argument that the nakshatra lists begin with Krittika, the Pleiades, and therefore date to the third millennium BCE was made by B. G. Tilak in The Orion (1893), and lately by Subhash Kak. It is an inference. The texts do not make it, and most historians of science do not accept it.

The Greek hinge

Around the beginning of the Common Era, something arrives.

The founding surviving text of Indian horoscopy is the Yavanajataka, and the title gives the game away before you open it. Yavana is the Sanskrit for Greek, from “Ionian.” Jataka means nativity. Nativity according to the Greeks.

David Pingree, in his 1978 Harvard edition, reconstructed the chain: a Greek prose text Sanskritised by one Yavaneshvara around 149 or 150 CE, then versified by Sphujidhvaja around 269 or 270. That dating entered every textbook, and it is no longer secure. Bill Mak, in 2013, working from a previously unreported manuscript, showed that Pingree’s key readings in the final chapter, the date among them, rest on Pingree’s own emendations rather than on what the manuscripts say, and revised the text to the fourth to sixth century CE. Notice what the revision does. It makes the text later, and no more indigenous. Both datings put it centuries after Lagadha, and squarely after sustained Greek contact. The date is disputed. The direction of borrowing is not.

The vocabulary settles it from the inside. Greek hora becomes Sanskrit hora, the hour, the ascendant, and by extension the whole natal art, horashastra. Kentron becomes kendra, the angular houses. Trigonon becomes trikona, the trines. Dekanos becomes drekkana; lepton becomes lipta, a minute of arc. Even the sign-names in the Yavanajataka are Greek in transliteration before anyone translated them into the Sanskrit still used: Kriya for Krios, Tavuru for Tauros, Jituma for Didymoi.

The tradition’s own greatest authority did not bother with a disguise. Varahamihira, the master astronomer of sixth-century Ujjain, wrote this into his encyclopaedia:

The Greeks are barbarians, yet this science is well established among them; they are honoured as though they were sages. How much more, then, a brahmin who knows the science of the stars.

Brihat Samhita 2.15, in M. Ramakrishna Bhat’s translation

Fifteen hundred years ago a brahmin astronomer told his readers exactly where the horoscope came from, and said it was good. It is his descendants who have forgotten.

What the astrology actually is

Describe the system as it stands, because it is elegant, and most of its critics have never looked at it.

Twelve rashis, signs, of thirty degrees each. The lagna, the ascendant, is the sign rising at the eastern horizon at the moment of birth, and fixing it is the whole reason the hour matters. Twelve bhavas, houses, counted from the lagna, governing the departments of a life. The kendras (houses 1, 4, 7, 10) are the angles, the trikonas (1, 5, 9) the trines. Note the two Greek loans sitting in that sentence.

Nine grahas, and here a word must be rescued. Graha comes from the root grah, to seize. A graha is a seizer, and it is not a planet. The nine are the sun, the moon, Mars, Mercury, Jupiter, Venus, Saturn, and then Rahu and Ketu, the chhaya-grahas, the shadow-grahas, which are the ascending and descending nodes of the moon’s orbit: the two points where eclipses can happen. They are geometry, personified by the tradition as the severed head and body of a demon. So of the famous nine planets, one is a star, one is a satellite, two are mathematical points, and the earth is missing.

Then the dashas, and here Indian astrology stops borrowing and becomes itself. A dasha is a planetary period, a timing engine that says when the promise in a chart will fire. Vimshottari dasha runs a hundred and twenty years apportioned among the nine grahas, its starting point fixed by the moon’s nakshatra at birth. It has no Greek parallel. Western astrology times by transit and progression; Indian astrology times by dasha. That is a real indigenous difference, and it deserves saying as loudly as the loanwords.

The dashas are codified in the Brihat Parashara Hora Shastra, and about that book honesty is required. Tradition ascribes the working bible of modern Indian astrology to the ancient sage Parashara, father of Vyasa. Scholarship does not: Pingree treats it as pseudepigraphic and early-medieval, Jan Gonda places its first part after 600 CE, and the printed vulgate rests on late manuscripts with heavy interpolation. Its authority is real. Its claimed antiquity is not.

Two more arts sit beside the birth-chart: muhurta, the choosing of an auspicious moment for an act, and prashna, horary, which reads the chart of the moment a question is asked rather than the moment a person was born.

Why your sign is different

This is the most useful technical thing anyone can hand a reader, and it is the question people actually ask.

Western astrology uses the tropical zodiac, in which zero degrees Aries is defined as the point of the spring equinox. Indian astrology uses the sidereal zodiac, anchored to the fixed stars.

The earth’s axis wobbles, a precession completing a circuit in about 26,000 years, drifting some fifty seconds of arc a year, about one degree every seventy-two. So the equinox slides against the stars. The two zodiacs coincided in the third or fourth century CE and have been separating ever since. The gap between them is the ayanamsha, and today it is about twenty-four degrees.

That is nearly a whole sign. And that, and nothing else, is why you are a Leo in a Western horoscope and a Cancer in an Indian one. There is no contradiction: the two systems measure from different zeroes, and never claimed otherwise. It is arithmetic, not mysticism.

One thing more, quietly. Indians do not agree about the ayanamsha either. The government standard is the Lahiri, anchored so that the star Chitra (Spica) sits at 180 degrees, adopted on the recommendation of the Calendar Reform Committee chaired by the physicist Meghnad Saha, which reported in 1955. But B. V. Raman’s ayanamsha and the Krishnamurti ayanamsha differ from Lahiri’s by degrees. A system that cannot agree on its own zero point is worth noticing.

The traffic went both ways

The Greek debt is real. It would still be false to leave India in the role of borrower.

The older question is Mesopotamia. Pingree argued in 1973 that Babylonian material reached India through Iranian intermediaries, pointing to the 3:2 ratio of longest day to shortest, the linear zigzag function for daylight, and the water clock, all Mesopotamian commonplaces. The Indian counter-case is serious and technical. Yukio Ohashi calls Pingree’s derivation “definitely wrong” and argues the field grew out of Indian observation, and the IIT Bombay computation finds the text’s own daylight data fitting Indian latitudes. The parallels are real; the direction of borrowing is unresolved. The Greek case is different in kind, because loanwords cannot be explained away and structural similarity can.

Then the return journey. Around 770 CE an Indian scholar reached the court of the Abbasid caliph al-Mansur in Baghdad carrying a Sanskrit siddhanta, and from it al-Fazari and Yaqub ibn Tariq produced the Zij al-Sindhind, Indian astronomy in Arabic, its mean motions taken from Brahmagupta. Al-Khwarizmi wrote his own around 820 on Indian parameters, and along the same road travelled the Indian decimal numerals.

And the sine. Indian astronomers replaced the Greek chord with the half-chord, ardha-jya, shortened to jya, a bowstring. Arabic transliterated it as jiba, written without vowels. Later readers, meeting the bare consonants, read the word as jayb, which means a bosom, a bay, a fold, and the twelfth-century Latin translators rendered jayb faithfully as sinus.

The English word “sine” is a mistranslation of a Sanskrit word for a bowstring. It has sat in every trigonometry textbook in the world for eight hundred years.

The traffic came back a third time. From the thirteenth century Indian astrologers absorbed a Perso-Arabic system of annual prediction, the Tajika school, named from the Persian for an Arab, whose technical vocabulary descends from Sahl ibn Bishr, as Martin Gansten has shown. Most Indian astrologers casting an annual chart today have no idea the machinery is Arabic.

Madhava’s series, and the trap

The greatest thing that ever came out of jyotisha has nothing to do with fortune-telling.

Around 1350, at Sangamagrama in Kerala, a mathematician named Madhava found the infinite series: power series for the arctangent, the sine and the cosine, and from the arctangent series a series for pi, with rational correction terms to speed its convergence. In Western nomenclature these are the Gregory, Leibniz and Newton series. Madhava precedes them by roughly two hundred and fifty to three hundred years. This is not a nationalist talking point. It is the settled position of historians of mathematics.

None of Madhava’s own works survive intact. We know the results are his because his successors name him. Nilakantha Somayaji’s Tantrasangraha (1500) attributes them to him, and Jyeshthadeva’s Yuktibhasha, written around 1530 in Malayalam, gives the derivations, the yukti, which is why it is sometimes called the first text of calculus. The recovery of all this is overwhelmingly the work of an Indian scholar, K. V. Sarma, whose History of the Kerala School of Hindu Astronomy (1972) put the school back on the map. It had been announced once and disbelieved: in 1834 an East India Company civil servant, Charles Whish, told the Royal Asiatic Society that Indian mathematicians had the series for pi before Newton was born, and was ignored for a century.

Nilakantha did one thing more. In the Tantrasangraha he set Mercury, Venus, Mars, Jupiter and Saturn orbiting the sun, while the sun orbits the earth. State it precisely, because it is routinely inflated. That is a geo-heliocentric model, reached independently, roughly a century before Tycho Brahe proposed the same in 1588. It is not Copernicus. And no evidence exists that any of it reached Europe. That transmission has been proposed. It remains a hypothesis without a document.

Now the trap, and it must be named, because it is why this article exists.

The commonest defence of astrology in India is to point at Aryabhata’s spinning earth and Madhava’s series and say: look, this is a science. It is a bad argument, and it serves nobody. Those achievements belong to ganita and gola, calculation and spherics, which are the astronomy. They do not certify the jataka. The tradition itself divides jyotisha into three branches, ganita (computation), samhita (omens) and hora (the natal art), and it is a category error to let the first vouch for the third. The sharpest critic of this move is Meera Nanda, in Prophets Facing Backward (2003), who is also polemical and contested, and both things are true at once.

The astronomy is not vindicated by being called astrology. The astrology is not vindicated by having astronomy in the family.

The test

Astrology makes claims that can be checked. They have been checked.

The strongest study is Indian. In Current Science 96.5, dated 10 March 2009, the astrophysicist Jayant Narlikar, the statistician Sudhakar Kunte, and the rationalists Narendra Dabholkar and Prakash Ghatpande published “A statistical test of astrology.” They took a hundred birth charts of intellectually bright school students and a hundred of mentally handicapped children from Maharashtra, cast in the standard Indian manner, and asked volunteer astrologers to sort them. A panel of astrologers had agreed in advance on the chart-signatures to look for; the astrologers helped design the test they were sitting. Twenty-seven returned results, each judging forty charts.

Their success rate came out marginally below chance.

Indian astrologers, Indian charts, Indian methods, Indian scientists, and a protocol the astrologers themselves helped write. It is the least dismissible experiment in the literature.

It is not the one usually cited. Shawn Carlson’s double-blind test in Nature (1985) found twenty-eight astrologers doing no better than chance at matching personality profiles to charts, and it is routinely presented as a knockout. It is not one. Suitbert Ertel re-analysed Carlson’s data in 2009 and argued that on the two tests Carlson himself regarded as valid, pooled, the astrologers scored above chance at marginal significance. That challenge is real, and it is why the Indian study is the better anchor: no comparable blow has landed on it.

The tradition has a reply, and it is not stupid. In the classical framing, jyotisha is no prediction machine. It is a karma-darshana, a way of seeing the ripening of prarabdha karma, that portion of past action which has begun to bear fruit in this life, a doctrine we set out in our account of karma and rebirth. The Brihat Parashara Hora Shastra presents the grahas as instruments of a karmic reckoning rather than as causes. On that reading, a laboratory sorting charts into bright and handicapped is testing a claim jyotisha never made. Whether that rescues the practice or merely moves it beyond the reach of evidence, the reader must settle. It is worth knowing that the tradition makes the argument.

There is a public chapter to this. In February 2001 the University Grants Commission funded degree courses in Jyotir Vigyan, “the science of Vedic astrology,” at Indian universities. That July more than a hundred scientists, including Narlikar and Yash Pal, himself a former UGC chairman, signed a letter of protest. Their objection was careful: to calling it vigyan, science, and funding it from a science budget, since its criteria of validation are not those of science. The other side has a case too. Jyotisha is a classical shastra with a vast literature, and universities teach Nyaya and Mimamsa without endorsing their metaphysics.

The biologist P. M. Bhargava litigated, and lost. The Andhra Pradesh High Court dismissed the writ; the Supreme Court dismissed the appeal in May 2004. In 2011 the Bombay High Court dismissed a separate petition against astrologers under the Drugs and Magic Remedies Act, relying on a government affidavit describing astrology as a trusted science. Be precise, because this is misreported in both directions. Neither court ruled that astrology is a science. One declined to interfere with a university’s control of its own curriculum. The other decided the scope of a statute on the strength of a drug-controller’s affidavit.

Eighty-three per cent

Here is the number that ends the argument, and neither side quotes it.

The Pew Research Center’s Religion in India survey (2021), a face-to-face study of about thirty thousand Indian adults, found that 44 per cent of Indians believe in astrology, and 49 per cent of Hindus. That is the figure the debate fights over.

The other figure is 83 per cent. That is the share of Indians who say they fix important dates by auspicious timing.

Read them together and the real subject appears. Belief in astrology is contested in India. The calendar is not. Even Indians who do not believe in astrology schedule their weddings, their housewarmings and their journeys by the panchanga, the five-limbed almanac of tithi, vara, nakshatra, yoga and karana that we unpack in our chapter on reading the Hindu calendar. It is still printed annually, still sold at temple gates, still the arbiter of when Diwali falls, and because regional almanacs compute differently, it is why the same festival lands on different days in different states.

That is the register in which most Indians meet jyotisha, and it has little to do with prediction. It is permission. Is this a good time? The kula-jyotishi, the family astrologer, is a real social role, part counsellor, part genealogist, part scheduler. Contempt for a billion people asking for a good day is contempt for the reader.

The harms belong in the same breath, and they are real.

Mangala dosha, the manglik affliction, declares a person dangerous to their future spouse if Mars sits in the 1st, 2nd, 4th, 7th, 8th or 12th house. That is six houses out of twelve. A defect that catches close to half the population by construction is not a diagnosis. It is a market. Its consequences in lived India are engagements broken, marriage prospects narrowed, particularly for women, and the remedy of kumbha vivaha, a symbolic marriage to a clay pot or a banana tree, performed to absorb the flaw before the real wedding.

The guna milan score, thirty-six points across eight kootas, is a standard field on every Indian matrimonial site, computed before the photographs are exchanged. Two components deserve a hard look. The varna koota grades compatibility on an explicitly caste-derived scale. And nadi, the heaviest at eight points, is routinely sold today as a check on genetic compatibility. It is nothing of the kind, and there is no basis for the claim whatsoever. It is a modern retrofit in a lab coat, and it descends from the Brihat Parashara Hora Shastra, not from the Veda.

Beneath all of it runs the remedy economy: gemstones, yantras, dosha-removal packages, industrialised now by apps and television astrologers, largely unregulated. The mechanism never changes. Misfortune is reframed as an astral debt, and the debt can be bought off.

Most practising astrologers are not predators. That should be said too.

What is left when you have said all this

A ritual religion needed to know when, and out of that need came three thousand years of continuous astronomical work. Lagadha counting days on the Kurukshetra plain. Aryabhata setting the earth turning and being overruled for it. Varahamihira admitting in writing where the horoscope came from. Brahmagupta’s numerals travelling to Baghdad. Madhava finding the infinite series in Kerala. Shankara Balakrishna Dikshit, a Marathi schoolteacher, writing the first modern critical history of the whole tradition in 1896, while his country was being told it had never had a science at all. That record needs no horoscope to justify it, and it is the direct offspring of the Veda’s need to be recited at the right hour.

The astrology is a different thing. Borrowed, brilliant, twice foreign, thoroughly naturalised, and it does not do what it says it does. The tradition can survive being told that. It has survived worse.

Aryabhata was right, and he lost.

The sky did not notice either way.