Grahan Dosha, Honestly Explained

Reviewed by Pt. Amit Pandey · September 2026

You were probably told that your Sun or Moon sitting beside Rahu or Ketu means you were born under an eclipse. This page opens the chapter of the Brihat Parashara Hora Shastra that actually deals with eclipse births, quotes it with its printed pages, and shows that it describes something else entirely — then measures how often each thing really happens.

The popular rule is simple: if the Sun or the Moon shares a sign with Rahu or Ketu in your birth chart, you have Grahan Dosha — the eclipse affliction. Sun with a node is called Surya Grahan Dosha, Moon with a node Chandra Grahan Dosha.

It is not a rare finding. Among the 77 published celebrity charts on this site, 21 carry it — better than one chart in four. Hold on to that number; it turns out to be most of the story.

The first thing to know

The classical eclipse verse is about the sky on the day you were born — not about your birth chart.

Brihat Parashara Hora Shastra has a chapter dedicated to exactly this subject. It is Chapter 91, "Remedies from Birth in Eclipses", printed pp. 1023–1025 in G. C. Kapoor's Volume 2. Its condition is a solar or lunar eclipse actually taking place at the moment of birth — an event in the sky on one date, the kind of thing a calendar records.

That is a different condition from a luminary sharing a sign with a node, and the difference is not a technicality. We show below exactly how different, in numbers.

One more thing worth saying plainly: the phrase "Grahan Dosha" appears in none of the classical sources we have read — not in BPHS in either of the two independent translations we checked, not in Phaladeepika, not in Saravali. The chapter's own Sanskrit title is ग्रहणजातशान्ति, "the pacification for one born in an eclipse." It is not the name of a dosha.

What Parashara actually wrote

Here is the chapter's opening in Kapoor's translation, printed p. 1024. He renders the run of verses as a single block, so this covers slokas 1 to 14:

> "1-14. The Sage said—O Brahmin! a person whose birth takes place at the time of solar or lunar eclipse, suffers from ailments, distress and poverty and faces danger of death. Therefore I am going to describe for the benefit of the mankind the remedial measures required to be undertaken to nullify the above evil effects."

And the Sanskrit of verse 1, printed p. 1023:

> सूर्येन्दुग्रहणे काले येषां जन्म भवेद् द्विज । व्याधिः कष्टं च दारिद्र्यं तेषां मृत्युभयं भवेत् ॥१॥

We checked the Devanagari for one specific reason: to see whether "at the time of" was the translator's addition. It is not. **काले — kāle, "at the time" — is in Parashara's own verse.** The temporal condition belongs to the text, not to its English.

The rest of the line maps cleanly: vyādhi → ailments, kaṣṭa → distress, dāridrya → poverty, mṛtyubhaya → danger of death. Nothing was added and nothing was dropped.

And a second translator, independently. Girish Chand Sharma numbers the same chapter 93 rather than 91 — the two editions run two chapters apart throughout — and titles it "Propitiation of Planets for Births During Eclipses". His rendering of the same verse, printed p. 761:

> "1-4: Maharishi Parasara said — O Vipra! Whosoever is born during the solar or lunar eclipses runs the risk of sickness, sufferings, poverty and death. Therefore, for the benefit of mankind, I wish to narrate the method of propitiation for the alleviation of such afflictions."

Two translators, working from the same Sanskrit, both read it as an event: born during an eclipse. Neither reads it as a pattern in a chart.

The chapter's own remedy settles the question

If there were any doubt about whether Chapter 91 means a real eclipse or a chart pattern, its remedy resolves it. The first instruction, printed p. 1024, is to prepare an idol:

> "(a) in gold of the deity of the nakshatra in which the eclipse takes place"

and again in the list of offerings, printed p. 1025:

> "(d) white flowers for the deity of the nakshatra in which the eclipse takes place"

The nakshatra in which the eclipse takes place only exists if an eclipse took place. It is a fact about a particular day. You cannot derive it from the Sun sitting in the same sign as Rahu, because that arrangement holds for weeks at a stretch and contains no eclipse in the overwhelming majority of them.

The chapter's neighbours point the same way. Chapter 85, "Inauspicious Births", printed p. 1009, is the preamble governing this whole block, and it lists the moments it has in mind:

> "…if they take place on Amavasya (the last day of the dark half of the month), on Chaturdasi (14th tithi) Krishna Paksha…, in Bhadra Karana, in the nakshatra of the brother, in the nakshatras of father and mother, at the time of entry of the Sun in a sign (संक्रान्ति), the time of Pata (पात), at the time of solar and lunar eclipses, at the time of Vyatipata (व्यतीपात), in gandantas of all the three kinds…"

Every item on that list is a calendar event — a tithi, a karana, a solar ingress, an eclipse. Not one is a planetary conjunction in a natal chart. The eclipse is keeping the company it was always meant to keep.

Two different conditions, side by side

The verse and the popular rule are usually presented as the same thing. They are not, and the gap is easiest to see laid out.

How often each one actually happens

This is where the two conditions separate decisively, and the arithmetic is simple enough to check yourself.

The popular rule. Rahu and Ketu are always exactly opposite one another, so between them they occupy 2 of the 12 signs. The chance the Sun lands in one of those is about 1 in 6. The same for the Moon. The chance that at least one of them does is therefore roughly 1 − (10/12)² ≈ 30.6%.

What do we observe? Of the 77 published charts on this site, 21 carry the configuration — 27.3%. That is the arithmetic, reproduced. And the agreement is the finding: a rule that fires on close to a third of all charts is not telling you something about you. It is telling you how many signs the nodes occupy.

The verse's condition. We measured this rather than estimating it. Using the ephemeris the site already runs on, we sampled 3,000 random instants between 1850 and 2010 and asked whether an eclipse was in progress:

- a solar eclipse under way somewhere on Earth: 0.17% - a lunar eclipse in its umbral phase: 0.03% - both together: 0.20% — or 0.37% if penumbral lunar eclipses are counted

So the popular rule fires somewhere between seventy and a hundred and fifty times more often than the condition the verse actually names.

What our own charts show

0 of 66

We ran every published chart with usable birth data — 66 of the 77, the rest missing a recorded time or timezone — against the real eclipse record. Not one was born during a solar or lunar eclipse. That includes all 16 of the flagged charts whose birth data we could use.

We want to be careful about what that does and does not prove. At a base rate of 0.2%, the expected number of eclipse births among 66 people is about 0.13. Finding zero is exactly what chance predicts. It is not evidence that the verse is wrong, and a larger group would eventually turn one up.

What it does show is more modest and more useful: across every chart we publish, there is not one person to whom Chapter 91's verse actually applies — while a quarter of them are being told they have "Grahan Dosha".

(The detector was checked against five eclipses known to have happened — the total solar eclipses of 1999 over Munich, 2017 over Nashville and 1995 over Delhi, and the total lunar eclipses of 2018 and 2000 — and against control dates with none. It identified all seven correctly.)

Why the word “eclipse” is all over the classical texts anyway

If you go looking, you will find "eclipse" scattered through the classical literature, and it is easy to read that as confirmation. It is not. It is a translation habit.

Phaladeepika mentions eclipses twenty-six times in the edition we hold. Every one of them refers to combustion — a planet too close to the Sun to be seen, astaṅgata in Sanskrit — which English translators of this period routinely render as "eclipsed by the Sun's rays." Mantreśvara even gives the exact distances:

> "The Moon is said to be eclipsed if she is 12° away from the Sun. Mars is eclipsed when he is 17° away from the Sun… Jupiter, Venus and Saturn are said to be eclipsed if they are 11°, 10° and 15° respectively away from the Sun."

That is a rule about visibility and planetary strength. It has nothing to do with an eclipse of the Sun or Moon, and nothing to do with Rahu or Ketu. Saravali's handful of mentions are the same thing, or plain metaphor — "miserliness eclipses all the good attributes."

So: no passage in either text treats a luminary–node conjunction as an eclipse affliction. That is a bounded claim covering the volumes we hold, and one gap is worth naming — our Saravali is Volume 2 only; Volume 1 is not in our library.

The remedy question, answered honestly

This is what most people arrive here looking for, so here is the straight answer.

Chapter 91 does prescribe a remedy — a detailed one, running fourteen verses. Idols are prepared according to one's means: gold for the deity of the nakshatra in which the eclipse fell, gold for the Sun if the birth was during a solar eclipse, silver for the Moon if lunar, lead for Rahu. A patch of ground is smeared with cow dung and covered with new cloth. Offerings follow by colour — red for the Sun, white for the Moon, dark for Rahu — then mantras, then a fire offering with specified woods: aak for the Sun, palas for the Moon, doorva grass for Rahu.

And then, printed p. 1025, step 6:

> "Sprinkle the holy water of the Kalasha on the child born (and his parents)"

That is the difficulty. The rite is performed on an infant, by its parents, in the days after a birth. Kapoor scopes this entire block of chapters to a qualified priest in his note at Ch. 92, p. 1028. There is nothing in it an adult reader can perform for themselves, and it answers a condition — birth during an actual eclipse — that almost certainly does not describe you.

We describe the rite because you asked what the text says, and this is what it says. We do not sell a remedy for it, and we would be cautious of anyone who does: a paid grahan shanti offered for a chart pattern is being sold against a verse about a different event.

What is actually worth looking at

None of this means your chart has nothing to say. It means the label is doing no work, and something more specific sits underneath it.

If it is your Sun sharing a sign with Rahu, that exact configuration is the one the Pitra Dosha material is built on — a different chapter, different claims, set out in our Pitra Dosha guide. Note that the Sun in Leo or Aries is widely held to withstand the contact.

If it is your Moon sharing a sign with a node, the question worth asking is about the Moon's own condition — its sign, its dignity, whether benefics aspect it — rather than the node's presence. A Moon in Cancer or Taurus is not in the same position as a Moon in Scorpio, whatever else shares the sign.

If every planet falls between Rahu and Ketu, that is a different pattern entirely — see Kaal Sarp Dosha, where we make a similar point about a similar name.

A challenged placement is not a verdict about your life. It marks an area that asks for conscious attention, which is a very different thing from a misfortune you have been assigned.

Key takeaways

• BPHS Ch. 91 (pp. 1023–1025) is the classical chapter on eclipse births, and its condition is an eclipse actually occurring at the moment of birth.
• The Sanskrit confirms it: काले, “at the time”, is in Parashara's verse, not added by the translator.
• The chapter's own remedy requires the nakshatra in which the eclipse fell — a fact that exists only if an eclipse happened.
• The popular rule — Sun or Moon sharing a sign with a node — occurs on about 27% of charts, matching the arithmetic prediction of 30.6%. That is a base rate, not a diagnosis.
• Birth during a real eclipse occurs at about 0.2% of moments. None of our 66 charts with usable birth data qualifies.
• The phrase “Grahan Dosha” appears in none of the classical sources we have read, including two independent translations of the BPHS eclipse chapter.
• Phaladeepika's 26 mentions of “eclipse” are all about combustion — a planet near the Sun — not about eclipses.
• The text's own remedy is an infant rite performed by the parents under a priest. There is nothing in it for an adult reader to do.

Frequently asked questions

What is Grahan Dosha?

In popular usage it means the Sun or the Moon occupying the same sign as Rahu or Ketu in a birth chart — Surya Grahan Dosha for the Sun, Chandra Grahan Dosha for the Moon. The name itself is modern: it appears in none of the classical texts we hold, and the BPHS chapter on eclipse births describes a different condition entirely.

Does Grahan Dosha mean I was born during an eclipse?

Almost certainly not. Birth during an actual solar or lunar eclipse occurs at roughly 0.2% of all moments. The Sun or Moon sharing a sign with a node occurs on about 27% of charts. Of the 66 published charts here with usable birth data — including all 16 flagged ones — not one was born during a real eclipse.

What does BPHS say about birth during an eclipse?

Chapter 91, printed pp. 1023–1025, opens: “a person whose birth takes place at the time of solar or lunar eclipse, suffers from ailments, distress and poverty and faces danger of death.” The rest of the chapter is a remedial rite. The condition is an eclipse actually taking place at birth, not a chart pattern.

What are the remedies for Grahan Dosha?

BPHS Ch. 91 prescribes a detailed rite — gold and silver idols, coloured offerings, a fire offering with specified woods — but it is performed on the newborn child by its parents in the days after birth, and Kapoor scopes the whole block of chapters to a qualified priest. There is nothing in it an adult can perform for themselves, and it answers a condition that probably does not apply to you. We sell no remedy for it.

How do I check Grahan Dosha in my kundli?

Find the sign occupied by Rahu, and Ketu directly opposite it. If your Sun or your Moon sits in either of those two signs, the popular rule is satisfied. It will be satisfied for roughly one person in four, which is the main thing worth knowing about it.

Is Grahan Dosha cancelled if the Sun or Moon is strong?

That is the common modern cancellation: the Sun in Leo or Aries, or the Moon in Cancer or Taurus, is held to withstand the contact. It is a reasonable extension of ordinary dignity rules. We apply it, and we say plainly that it is not drawn from the eclipse chapter, which contains no cancellation clause of any kind.

Why do classical texts mention eclipses so often then?

Because English translators of this period render astaṅgata — combustion, a planet too near the Sun to be seen — as “eclipsed by the Sun's rays.” All 26 of Phaladeepika's mentions of “eclipse” are that. It is a translation habit, not a reference to eclipses.

Sources & method

Every classical sentence on this page was read in the translator's printed edition and is cited to its printed page: Brihat Parashara Hora Shastra Ch. 91 (G. C. Kapoor, Vol. 2, pp. 1023–1025) and Ch. 85 (p. 1009), the same eclipse chapter in a second independent translation as Ch. 93 (G. C. Sharma, Vol. 2, pp. 761–763), and Phaladeepika on combustion. Chapter boundaries were confirmed either side — Ch. 90 closes on p. 1022, Ch. 92 opens on p. 1026. Two gaps are worth naming: our Saravali is Volume 2 only, and we have not swept Brihat Jataka, Jataka Parijata or Prasnamarga for eclipse-birth material. The frequency figures come from the site's own ephemeris, sampled over 3,000 instants between 1850 and 2010, with the eclipse detector checked against five eclipses known to have occurred and two control dates.