The Week and the Fortnight
As in the case of the Vedic Indian, or perhaps borrowing or inheriting his methods, the Buddhists noted the passage of the days and kept count of them by the nightly changes in the phases of the moon. The regularity of the recurrence, especially of the full-moon and new moon, obviously struck the Buddhist observer, as it did his contemporaries in his homeland and elsewhere, and these events formed points of punctuation in his journey through the days, which might have otherwise been for him a mere alternating succession of darkness and light. Consequently, we have the idea of two fortnights (pakha) or half-months (addhmasa). Both together composing the lunar month of twenty-nine days, which was uncomfortably adjusting itself to the Buddhists‘ solar month of 30 days, arrived at by the division of the year into twelve equal parts. It is true that the Buddha used the 'week’ or seven days (sattana) as a convenient unit of time occasionally, as witnessed in the Vinaya rules for leave-taking from the Rains-Retreat (Vin.i, p.138) and the timing of his meditation in various places shortly after the Enlightenment, but the fortnight seems to have proved a more distinct and dependable (and therefore more frequently used) unit of time, with the natural phenomenon of the moon’s waxing and waning by which to be easily guided. The two fortnights are distinguished as the light or bright sector (sukha pakkha), i.e., the moonlit nights, and the dark sector (kanha pakkha), i.e., the moon’s waning period, the word pakkha being generally used in preference to addhamasa when the two fortnights are being spoken of in relation to each other. The distinction between the two fortnights seems to have been familiar to and used by the Buddhists as early as the Digha (D.iii, p.182) and the Anguttara (A.ii, p.19) Nikayas.
The days of the week or fortnight do not appear to have had any names attached to them, but they were numbered within each fortnight. Thus the fortnight beginning with the first day after the full-moon and waning to the new moon has its intermediate days known by ordinal numbers, e.g., the 1st, 2nd, 3rd day etc. of the dark fortnight. About half way through the fortnight was the eighth day (atthami) assuming some religious importance by the same reason, and then at the end of that fortnight was the fourteenth day (catuddasi) which is the new-moon day. The numbering re-begins for the bright fortnight from the day after the new moon with another eighth (atthami) about mid-way and ends on the fifteenth day. (panchadasi: M.i, p.20) or (pannarasi: Pvu.p.38) which is the full-moon day. This system of numbering seems to have been particularly used for the noting of dates for the observance of the Uposatha by both clergy and laity, as well as for the entry into the Rains-retreat, the Pavarana ceremony, the recitation of the Patimokkha by monks, and other religious rituals and sacred observances of the Buddhist community in general.
There is no direct statement as to which of the fortnights was taken as preceding the other, probably because two systems were in vogue in the Buddhist background, one of beginning the month with the dark fortnight and the other with the bright, but as Thomas (ERE.Vol. 12,p.1067) points out, the fact that the dark fortnight is usually mentioned first and that the months of retreat began a day after the full-moon (Vin.i, p.137 and VinA.v.p.1067) may be taken as suggesting that the full-moon-ending month was the more regular system used by the Buddhists, in which case, the kanha pakkha could be deemed to have been the first half of the Buddhist month.
- (Two Systems: Later Indian Astronomical lore indicates the use of two systems: the purnimantavadhi system with the month ending with the full-moon day, and the amantavadhi system with the month ending on the new moon. This tradition is, in all likelihood, very much older and was probably prevalent even in Buddhist times. The Buddhists, apparently, were aware of both systems and atthami are usually referred to as "of the fortnight" (pakkhassa while no such qualification is made for pannarasi (full-moon day) which merely passes as "the fifteenth day". Thus the fifteenth day would be understood to be the "fifteenth" calculating from the first day after the new moon, while the fourteenth (catuddassi) and the eight (atthimi) would be calculated in relation to the fortnight (pakkha, "of the fortnight") ).
An even more convincing piece of evidence for this conclusion is the myth of the Four Guardian Deities contained in the Anguttara-nikaya (A.i, p.142) which designs to stress the relative importance of the four sacred days of a month. The sutta under reference tries to impress on the devote that when there is virtue on earth, the gods are glad, learning that their numbers in heaven will be on the increase. For this reason, the Ministers of the Assembly of the Four Guardian Deities (or Four Great Kings: cattaro maharajano), the sons of these deities and the deities themselves survey the world of men in turn in order to report their findings to their subjects, the devas of the Tavatimsa heaven. According to the passage, the ministers survey the earth on the eighth day (atthami), the sons on the fourteenth day (catuddassi) and the four guardian deities themselves on the fifteenth day. (panchadasi) Now, there is no reason to think, as Rhys Davids (Buddhism, p.139f) and Bateson (ERE.Vol., p.836) do (an error also occurring in the PED), that the fourteenth and fifteenth day must be taken to mean the fourteenth day from the new moon in short months and fifteenth day in the long. It does not confirm to the logic of the above myth to suppose that the four guardian Deities, the most important of the three groups, survey the earth in the long month only, in which case, the full moon (panchadasi uposatha) day of the long month must be considered a more important full moon day than that of the short, an assumption that is not in any other part of the scriptures. As a matter of fact, the tradition, both literary and popular, seems to consider each full moon day as the most important day of the month, the new moon day second in importance and the two eighth days (atthami) third. The argument of the myth must rather suppose that these celestial emissaries visit the earth at sufficiently regular intervals, the most important visiting is on the last and most important day of the month, the myth being built on the analogy of a royal court and its envoys responsible to its subjects. It is unlikely that it was imagined that the highest dignitaries visited the earth just one day after their sons had visited it or irregularly, only in " the Long months", for which concept too, in fact there is no traceable evidence in the texts.
It is legitimate, therefore to infer that the days of the three visits are given, not only in order of their importance but also according to the order in which they occurred within the month, and if so, to conclude that the month began on the first day after the full-moon, advanced to the first eighth, then to the new moon, again to the second eighth and ended with the full-moon night, the dark fortnight then preceding the bright The calendar-month then, lunar based as it is, would consist of four "weeks" of a varying number of days, the first and the third being each of eighth days duration while the second week was comprised of only six days and the fourth of seven. In fact, the Buddhist lunar ‘week’ as the period from one Uposatha to the other, does not necessarily correspond with the seven-day unit of time (sattaha) frequently utilized by the Buddha in the framing of Vinaya rules (Vin.i,p.138) and the regulation of his own activities (Vin.i,p.4).
The chart below gives us a view of the placing of the weeks and fortnights within the months, for one season of four months. However, according to a tradition preserved by Buddhaghosa, an uposatha coming at the end of a fourteen-day period, i.e. catussdasi occurs only six times during a year of 12 months. According to him, they came in the third and seventh pakkha or fortnight of each of the three seasons, hemanta, gimha and vassana. Thus the remaining 18 are invariably ‘fifteenth day’ i.e., pannarasi uposatha. What is interesting to observe is these ‘fourteenth day’ uposatha which we have indicated above to be days of the new moon, occurs every other month alternating with the ‘fifteenth day’ new moon. Perhaps, inspite of the new moon days of both types, the term catuddasi came to be exclusively used for the new moon day, reserving the term pannarasi for the day of the full-moon (kankhavitarani, p.3).
VASSÂNA SEASON [Rains-retreat]
[Reckoned according to Purnimântâvadhi System
i.e. the month commencing on the first day after the full-moon].
* early vassa begings, ** later vassa begins,
¢ atthamî, $ câtuddasî, § pañcadasî
+ Kattikâ temâsini [pavâranâ]; early vassa ends.
++ Kattikâ câtumasini [[[komudi]]) ; later vassa ends.
 although according to the name catuddasi the new moon should fall on the fourteeth day, the tradition preserved by Buddhaghosa assures us that this happens only on in alternative months.