the classical texts
Suryaprajnapti: Jain Astronomy of Two Suns and Two Moons
Published 2026-11-04 | Live Pandit Ji
The Suryaprajnapti (सूर्यप्रज्ञप्ति) is a book of the Jain canon on the sun, the moon and the nakshatras. The Beawar edition read for this page counts it among the upangas, the subsidiary limbs of the canon, and records that some lists make it the fifth upanga and others the sixth. Its language, in Thibaut’s description, is Jaina Prakrit. What the two sources report from it is a model of the sky and a calendar; neither describes any horoscopy in it.
The book is read here at second hand. The main guide is G. Thibaut, then Principal of Benares College, whose study “On the Suryaprajnapti” came out in two parts in the Journal of the Asiatic Society of Bengal for 1880: Part I in No. 3, pages 107 to 127, and Part II in No. 4, from page 181. Thibaut built on A. Weber’s chapter-by-chapter summary of 1868 and chose instead to group the doctrines by subject. He leaned more on the commentary of Malayagiri than on the text, because the manuscripts he found gave the commentary in full and quoted the text only by its opening words. The second source, used only for how the book is divided, is the Hindi edition with the original text from the Agam Prakashan Samiti of Beawar in Rajasthan, number 29 in its Jinagam series, edited by Muni Kanhaiyalal “Kamal”; the copy read is the second edition, whose title page gives Vira Nirvana Samvat 2522, Vikrama Samvat 2052 and June 1995.
20 Prabhritas: How the Book Is Divided
Both sources give the book twenty divisions called prabhritas, pahuda in the Prakrit. Only three of them are split further into prabhrita-prabhritas: the first into 8, the second into 3 and the tenth into 22, 33 sections in all, while the other seventeen run on without a break. Thibaut’s references to the order of the nakshatras and to where each stands beside the Moon come from that much-divided tenth prabhrita (X.1 and X.11 in his notation). The Beawar introduction counts 108 prose sutras and 103 verse gathas, and reckons the surviving text at about 2,200 shlokas in length. Its editor also guesses that the Suryaprajnapti and its companion, the Chandraprajnapti, first circulated under one joint name and were only later counted as two upangas; the Beawar volume prints them as a pair.
Why Two Suns and Two Moons?
The doctrine outsiders noticed first is the doubling, and Thibaut points out that Colebrooke’s 1807 account of the Jains described little else: two suns, two moons and a double set of nakshatras. Taking up a hint from Colebrooke, he traces it to a plain difficulty. In the old Indian picture the sun circles Mount Meru, and the Jain authors gave that circle four quarters, one for each direction, each quarter bringing day to the lands on its side of the mountain. If the sun spends equal time in every quarter, it ought to return to Bharatavarsha, the southern segment of the world, only after a whole circle; yet it rises there every morning. The Suryaprajnapti’s answer is that the sun rising today is not the one that set last night. Two identical suns travel the same circle half a turn apart, and each is seen over Bharatavarsha on alternate days. The same argument supplies two moons and two sets of stars.
Thibaut’s account calls the pair the Bharata sun and the Airavata sun. At sunrise on the first day of the yuga the Bharata sun stands at the south-east point of its circle and is seen rising in Bharatavarsha, while its twin, at the north-west point opposite, rises for the lands north of Meru. By evening each has gone a quarter further, the Bharata sun bringing day to the lands west of Meru and the Airavata sun to those east of it. Next morning the two have changed places, and on the third morning each is back where it started.
In the working, Thibaut adds, the doubling changes almost nothing. When one sun has run through all twenty-eight nakshatras it enters a second Ashvini, which has the same name, size and neighbours as the first, while its twin enters the first one. A reader can therefore follow the text as though there were one of each, and the rest of this page does so.
Meru, Jambudvipa and the Sun’s 184 Circles
The stage is the flat earth of the Puranas, as Thibaut summarises it: concentric rings of land called dvipas, separated by rings of ocean, with Mount Meru at the centre. The first ring is Jambudvipa, 100,000 yojanas across, with the salt ocean around it; Bharatavarsha lies in its southern segment, Airavata in the northern, and Videha to the east and west of Meru. The suns move round Meru at a fixed height of 800 yojanas and the moons 80 yojanas higher, at 880. Rising and setting are not real events in this model. A sun is said to rise for a place when it comes into view there and to set when it passes out of sight.
The sun’s yearly swing north and south becomes a change of circle. On the longest day the sun runs the innermost circle, 99,640 yojanas across, 49,820 from the centre of Meru. Each day after that it moves to a wider circle, 2 48/61 yojanas further out, until at the winter solstice, 183 days on, it runs the outermost, 100,660 yojanas across; then it works back inward by the same steps. That makes 184 circles in a band 510 yojanas wide, the innermost 180 yojanas inside the edge of Jambudvipa and the outermost 330 yojanas out over the salt ocean. All the circles are at one height. The sun looks lower in winter only because it is further away. The text even asks how the sun passes from one circle to the next, and on Thibaut’s reading it prefers a continuous spiral to a jump.
Day length moves in step. As in the Vedanga Jyotisha, the longest day is 18 muhurtas and the shortest 12, out of thirty in a day and night. Spread evenly over 183 days, the change comes to 2/61 of a muhurta a day; Thibaut calls the even step a mistake. The table sets the two extreme circles side by side.
| Circle | Across (yojanas) | Round (yojanas) | Yojanas per muhurta | Day (muhurtas) | Share lit by each sun |
|---|---|---|---|---|---|
| Innermost (longest day) | 99,640 | 315,089 | 5,251 29/60 | 18 | 3/10 |
| Outermost (shortest day) | 100,660 | 318,315 | 5,305 15/60 | 12 | 2/10 |
The circumferences are not three times the diameter. The Suryaprajnapti multiplies a diameter by the square root of ten, which gives Jambudvipa a circumference of 316,227 yojanas and some smaller units, and the table applies the same rule to the two circles. Each sun covers half its circle in the thirty muhurtas of one day and night, so a whole circle takes sixty, and the speed is the circumference divided by sixty. On the longest day the sun is up for 18 muhurtas and needs 9 of them to reach the meridian, so the text puts the distance at which it first comes into view at 47,263 yojanas and a fraction, measured along its circle. The last column follows the same day lengths: on the longest day the two suns together light three fifths of Jambudvipa, and on the shortest two fifths.
Older Opinions the Text Records
The Suryaprajnapti does not stop at its own figures. It reports what other teachers held, and Thibaut takes some of those views to be older than the book. Three schools made the innermost circle 1,133, 1,134 or 1,135 yojanas across, with a circumference of 3,399, 3,402 or 3,405, each exactly three times its diameter. Thibaut reads that rough ratio, set against the book’s own square root of ten and its far larger Jambudvipa, as a mark of greater age. Other teachers had the sun cover a fixed 6,000, 5,000 or 4,000 yojanas a muhurta whatever its circle; on the first figure the stretch it lights is 108,000 yojanas on the longest day and 72,000 on the shortest. Still others let its speed change through the day. The book also records views on what the sun is, among them that it is no deity at all but a mass of rays that gathers in the east each morning and breaks up in the west each evening.
The Moon’s 15 Circles and the Count of 1,830
Thibaut observes that the greater part of the Suryaprajnapti concerns the Moon, yet it never sets out her motion in one place, so he rebuilds it on the pattern of the sun’s. The Moon too circles Meru, more slowly than the sun, and the book measures each kind of body by how many daily circles it completes in one five-year yuga.
| Body | Daily circles in one yuga |
|---|---|
| The nakshatras | 1,835 |
| The sun | 1,830 |
| The Moon | 1,768 |
The gaps between those numbers are the yuga’s other counts. The nakshatras gain 5 circles on the sun, its five trips through them; they gain 67 on the Moon, her sidereal revolutions; and the sun gains 62 on the Moon, the number of times the two meet, which is the count of synodic months. With two suns and two moons, each sun runs 915 of those circles and each moon 884. The Moon’s own swing north and south lasts one nakshatra month, and the book gives her only 15 circles against the sun’s 184. Thibaut adds that it knows nothing of the tilt of the Moon’s path, although it had noticed that she passes north of some nakshatras at one time and straight through them at another.
Five Years, 1,830 Days
The calendar behind all these counts is the five-year yuga. In Thibaut’s summary it holds five solar years of 366 days, each split into two ayanas of 183 days and into twelve solar months of thirty and a half days. Beside these runs a civil count that the book calls the karma year, 360 days long, so every solar year exceeds it by 6 days, the atiratras. The 62 synodic months of the yuga last as long as 61 civil months of thirty days, which makes 62 tithis equal to 61 days; the count therefore drops one tithi at every sixty-second. The months are grouped into five lunar years of 12, 12, 13, 12 and 13 months, and the third and fifth, the thirteen-month ones, are called abhivardhita, increased. Thibaut infers that the 31st and 62nd lunations were never counted on their own but merged with the month that followed, so the extra month needed no name. He found no trace in this book, nor in the Vedanga, of the correction for the over-long year that Colebrooke had supposed the Vedanga to carry.
The two books part company at the starting point. The Vedanga Jyotisha and Garga open the yuga at the winter solstice, with a new moon and the Sun and Moon together at the start of Dhanishtha. The Suryaprajnapti opens it at the summer solstice, with a full moon at the start of Abhijit and the sun opposite, in Pushya. Thibaut reads the change as one place where the author put what he saw above the old authorities, since by his time the winter solstice had moved too far from Dhanishtha to be ignored. He also tried a rough measure of the gap: Shravana and Abhijit together come to about 17.3 degrees, and at 72 years to the degree of precession that would put some 1,246 years between the two solstice records. He declined to date the Suryaprajnapti from it, because, as Whitney had shown, the Vedanga’s own date could not be fixed. This page gives no date either.
28 Nakshatras of Unequal Size
The book always counts twenty-eight nakshatras, Abhijit among them, and never makes them equal. It sorts them by how long the Moon stays with each. Thibaut shows that the sun’s time with each comes out of the same figures: write the Moon’s time in sixty-sevenths of a muhurta, then put five where sixty-seven stood, because the sun goes round five times in the yuga while the Moon goes round 67.
| Moon stays | Sun stays | How many | Nakshatras |
|---|---|---|---|
| one day and night (30 muhurtas) | 13 days 12 muhurtas | 15 | Revati, Ashwini, Krittika, Mrigashira, Pushya, Magha, Purva Phalguni, Hasta, Chitra, Anuradha, Mula, Purva Ashadha, Shravana, Dhanishta and Purva Bhadrapada |
| half a day and night (15 muhurtas) | 6 days 21 muhurtas | 6 | Shatabhisha, Ashlesha, Bharani, Jyeshtha, Ardra and Swati |
| a day and night and a half (45 muhurtas) | 20 days 3 muhurtas | 6 | Uttara Ashadha, Uttara Phalguni, Uttara Bhadrapada, Punarvasu, Vishakha and Rohini |
| 9 and 27/67 muhurtas | 4 days 6 muhurtas | 1 | Abhijit |
The four groups fill the circle exactly. The Moon’s times add up to 819 and 27/67 muhurtas, or 27 days and 9 27/67 muhurtas, which is the book’s periodical month; the sun’s add up to exactly 366 days. The book uses that muhurta total as its measure of the whole circle, so a nakshatra is a stretch of so many muhurtas rather than of degrees. The book’s own list starts from Abhijit, where the full Moon stands as the yuga opens. It mentions five views on which nakshatra comes first and spells out only the one beginning from Krittika, and Thibaut notes that it never names the equinoxes at all.
The unequal scheme did not die with the old calendar. Thibaut quotes Bhaskara’s Siddhanta Shiromani (Grahaganita, Spashtadhikara, verses 71 to 74), where the rough method of 27 equal parts is set aside for exact work, such as fixing processions and marriages: six nakshatras get one and a half portions, six get half, the other fifteen one each, and Abhijit takes what remains of the circle. Bhaskara’s own commentary names the two groups of six, and they are exactly the Suryaprajnapti’s long and short classes. Brahmagupta gives the same scheme in the 14th chapter of the Brahmasphuta Siddhanta, crediting it to the Paulisa, Romaka, Vasishtha, Saura and Paitamaha siddhantas and remarking that Aryabhata leaves it out. Thibaut took the way Abhijit is fitted in, as the excess of the sidereal month over 27 days, as a sign that it was added once that month was better known. How the sidereal month compares with the real Moon is worked on sidereal month vs synodic month.
Why the Siddhanta Writers Rejected Two Suns
The later astronomers mostly knew the Jain doctrine as something to refute. In the passage Thibaut quotes from Brahmagupta’s chapter of criticisms, the Dushanadhyaya, the objection takes one line: the teaching of the Jina, 54 nakshatras and two risings of the sun, is untrue, “since the revolution of the polar fish takes place within one day”. The figure 54 is the 27 nakshatras taken twice. Bhaskara makes the same point in the Siddhanta Shiromani, which Colebrooke translated: addressing the Jains, he asks how anyone who watches the polar fish go round can believe in doubled suns, moons and stars. Thibaut adds that Varahamihira’s Panchasiddhantika carries a passage to the same effect.
Neither passage spells out the step it rests on. Brahmagupta gives one reason, the turn of the polar fish within a single day, and Bhaskara points to the same sight; the conclusion both leave to the reader is that a sky which comes round in one day has no room for a second set of stars taking turns with the first.
Thibaut’s own verdict sits between the two camps. Once the sun is taken to circle Meru above a flat earth, he argues, two suns are an economical answer and alter very little in practice; what he faults is the minute detail into which the system was worked out without being checked against the sky, and the five-year yuga itself, which he says rests on “an utterly wrong assumption” about the year. For comparison he sets beside it the Chinese Tcheou-Pei, translated by Biot in 1841, where the sun likewise moves on concentric circles, seven of them round the pole, through 28 unequal star divisions. That book, however, counts 235 lunations in 19 years, and Thibaut rates it much the better of the two.
The Suryaprajnapti and the Vedanga Jyotisha
Setting the two side by side, Thibaut finds the rules for full and new moons and the like so close, often identical in method, that he treats the Suryaprajnapti as a fair witness to Indian astronomy before the siddhantas. He warns, though, that its rules cannot simply be carried across to explain the Vedanga’s obscure verses. The Vedanga has 27 equal nakshatras and starts from other solstice points, so the place of a given full or new moon can never come out the same in both. He suspected that the likeness ran deeper still, because verse 22 of the Vedanga speaks of the sun’s circles, surya-mandalani, which he took for daily circles round Meru of the Suryaprajnapti’s kind. The Vedanga’s own verses and its figures are read on the Vedanga Jyotisha page.
Related Pages
Where the siddhanta branch sits among the three branches of jyotish, and the other texts the site has read, is set out on the three skandhas of jyotish. The Jain festival year with its dates is on the Jain calendar for 2027. In Hindi: सूर्यप्रज्ञप्ति: दो सूर्य, दो चन्द्रमा.
Twenty-Seven Nakshatras, Counted by Sunrise
The Suryaprajnapti sizes its 28 nakshatras by the Moon’s muhurtas. The site’s nakshatra pages use the 27 equal nakshatras of the later reckoning and count how many of the 730 sunrises of 2026 and 2027 each one named.
Browse the 27 Nakshatras