Horā: three things with one name, and the one that is not an hour

A horā is one twelfth of the daylight, so it is sixty minutes only at the equinox. At this latitude it runs from about fifty-three minutes to about sixty-nine.

This page has to do something before it does anything else, because the word it is named after means three separate things and almost every page on the subject — including the version of this one that stood here from June 2022 — runs them together.

Three things called horā

The third one is the reason this matters commercially as well as doctrinally. When somebody searches "horā śāstra" they are usually looking for the natal branch, or for the text — and the most famous book in this subject has those words in its title. A page called Horā Śāstra that is entirely about planetary hours is answering a question that was not asked.

This rewrite is about the planetary hour, which is the useful, practical, daily thing. It says so in the title and it explains the other two so that nobody arriving from the wrong direction stays confused.

The three skandhas

The framework the old page left out, and the thing that makes the confusion make sense.

Jyotiṣa is classically divided into three skandhas — three trunks or branches.

→  Gaṇita, also called siddhānta — the mathematical and astronomical. Computing where the bodies are

→  Saṁhitā — the mundane. Weather, harvest, omens, the affairs of a place rather than a person

→  Horā — the natal. The chart of an individual and what is read from it

So horā, in its largest sense, is the branch this entire practice belongs to. The planetary hour borrows the same word, and the half-sign borrows it again. One term, three scales, and no page on the internet sorts them out.

A horā is not an hour

Now the practical part, and the correction.

The reckoning is straightforward and the old page had it right: take the daylight, from sunrise to sunset, and divide it into twelve. Take the night, from sunset to the next sunrise, and divide that into twelve. Twenty-four horās to a day.

Which means a horā is only sixty minutes at the equinox. It is a seasonal hour. Every other day of the year it is longer or shorter than sixty, and the further you are from the equator the more it moves.

The earlier version of this page said a horā has "a duration of approximately 60 minutes". That is the correction. I computed what it actually is.

At Dubai the day-horā varies by nearly sixteen minutes across the year. At Delhi by eighteen. At London by forty-four, which is to say a London day-horā in June is more than twice the length of one in December.

What that costs you

The error does not stay put. It accumulates, because each horā is measured from the end of the last one.

If you use a fixed sixty-minute clock from sunrise, then by the end of the twelfth horā you have counted twelve hours of daylight. The actual daylight is longer or shorter than that, and the gap is the drift.

At Dubai in June, somebody using clock hours is more than an hour and a half out by the end of the afternoon. That is not a rounding error. That is two whole horās — a completely different graha ruling the moment they act.

And it is worst in the afternoon, which is where most of the decisions are. The first horā of the day is right whichever way you count it, because both reckonings start at sunrise. The error grows with every horā after it.

Which graha rules which horā

The second correction, and this one is more interesting because the rule turns out to prove itself.

The horās cycle through the seven grahas in the Chaldean order — sorted by apparent motion, slowest first.

Śani, Guru, Maṅgala, Sūrya, Śukra, Budha, Chandra. Then back to Śani and round again, one graha per horā, through the day and through the night without a break.

The earlier version of this page gave the order the other way round — Chandra, Budha, Śukra, Maṅgala, Guru, Śani — with Sūrya set aside as an exception that sits at the head of the day. That is wrong, and the way you can tell is that it breaks the one thing this sequence is supposed to do.

The sequence generates the week

Here is the rule that makes the whole scheme hold together. The lord of the first horā after sunrise is the lord of the day. That is what makes Ravivāra Ravivāra: its first horā belongs to Sūrya.

So start on Ravivāra, walk twenty-four horās through the Chaldean order, and see where the next sunrise lands.

Twenty-four divided by seven leaves three. Three places along from Sūrya is Chandra — and the next day is Somavāra. Three on from Chandra is Maṅgala, giving Maṅgalavāra. Keep going and the entire week comes out in order, and on the seventh step you land back on Sūrya exactly.

I ran this mechanically to check it, and the Chaldean order reproduces the week exactly. The reversed order does not — it generates the week backwards, Ravivāra followed by Śanivāra followed by Śukravāra. Which is precisely why the old page needed an exception for Sūrya: the exception was patching a sequence that had been written the wrong way round. With the order stated correctly, no exception is required at all.

That is worth dwelling on for a moment, because it is a good demonstration of how this subject is supposed to work. The order is not a list somebody chose and you have to memorise. It is the order that makes the week come out right, and if you write it down wrong the week tells you.

Using a horā

The practical payload, and it is simpler than the tables that usually accompany it.

A horā carries the nature of its lord. So the question is not which activity appears on somebody’s list against which graha — it is what that graha signifies, and whether what you are about to do is that kind of thing.

Sūrya. Authority, the state, the father, standing. Dealings with government, applications to institutions, anything where a decision is being asked of somebody in a position of power.

Chandra. The public, the mother, movement, the mind at rest. Travel, matters concerning the home, conversations where somebody needs to feel comfortable rather than persuaded.

Maṅgala. Force, initiative, contest. Starting something that will need pushing, physical work, disputes and anything competitive. Not the horā for a delicate negotiation.

Budha. Communication, calculation, commerce, documents. Correspondence, accounts, contracts as instruments, teaching, anything involving words or numbers on paper.

Guru. Counsel, expansion, law, teachers. Advice sought or given, education, anything you want to grow, and the horā most traditions treat as the generally benign one.

Śukra. Relationship, value, beauty, negotiation. Pricing, terms, aesthetics, and dealings where the other party has to want the outcome as much as you do.

Śani. Structure, endurance, labour, the long term. Foundations, maintenance, anything meant to last, and work nobody will thank you for. Not the horā to launch something that needs momentum.

Work from the signification, not from a list. A list cannot anticipate what you are actually doing on a given Tuesday. The signification can, and it is the same reasoning a practitioner uses.

Where the scheme stops working

The old page raised this and it is worth keeping, because it is the honest limit of the method.

The horā divides the daylight. As you go north the daylight stretches and shrinks further, and the horā with it — at Reykjavík a day-horā runs from twenty minutes in December to more than a hundred and five in June. The unit is still defined but it has stopped being anything like an hour.

Go further, past the Arctic Circle, and on the days when the Sun does not set there is no sunset to divide at, and on the days when it does not rise there is no daylight at all. The scheme does not give a wrong answer there. It gives no answer.

That is not a failure of the tradition. It is a reminder of what the tradition is measuring — a real division of a real day, at latitudes where days behave the way they behaved where the method was built.

Where the word came from

One section on etymology, because the old page gave two accounts of it without noticing they cannot both be true.

The first is the traditional Indian derivation: horā is ahorātra — day and night — with the first and last syllables dropped. The second is the mainstream philological account, in which Sanskrit horā is a borrowing from Greek hōra, arriving with the Yavana material, and English "hour" descends separately from the same Greek word through Latin.

Those are different claims and this page is not going to pretend otherwise. The old version stated that English "hour" derives from horā, which is the one thing neither account supports — on the traditional derivation the two words are unrelated, and on the philological one they are cousins rather than parent and child.

I am naming the dispute rather than settling it. What I will say is that the arithmetic above does not depend on it in any way. Where the word came from is a question about language. What a horā is, is a question about the daylight, and that one has an answer you can check with a clock.

The practical version

Two things.

Do not use clock hours. Take today’s sunrise and sunset, subtract, divide by twelve. That is your day-horā, and any pañcāṅga or app worth using will already have done it. If the horās on a table are all sixty minutes long, the table is wrong for every day but two in the year.

And check the first one. The first horā after sunrise must belong to the lord of the day — Sūrya on Sunday, Chandra on Monday, and so on. If it does not, something in the reckoning has gone wrong, and you have found it in four seconds.