Outline of European Christian and Indonesian Muslim calendars


For a fuller discussion of the history of Muslim calendars in Indonesia, see:
Ian Proudfoot, Old Muslim Calendars of Southeast Asia, Leiden: Brill, 2006. Handbuch der Orientalistik. III. Southeast Asia, ed. V. Lieberman, M.C. Ricklefs, D.K. Wyatt, vol.17. 135pp. Includes CD-ROM “Takwim: Javanese and Malay date conversions”.

AD (Anno Domini)

The Christian era dates notionally from the birth of Christ, and is expressed through the Julian and later the Gregorian calendars.  The Gregorian calendar became current in Catholic Europe in 1582 or soon after, achieving very wide use across Europe by 1752.  By the eighteenth century the Julian calendar lagged 10 days behind the Gregorian calendar.
The Gregorian year normally begins on 1 January, but especially in the period before 1600 there was a chaotic diversity of concurrent years in use even in the same locality.  The most common beginnings were 25 December (a nativitate), 1 January (a circumcisione,  or stilo communo), and 25 March (ab annunciatione).  By the end of the sixteenth century, the use of 1 January was widespread, though not in England, where general use of the year beginning in 25 March continued until the eighteenth century.
The day begins at midnight (though mariners' logs also began the day at noon).
 
A few useful dates for adoption of the Gregorian calendar are:
    region    reform    beginning of year
Rome, much of Italy (except Savoy, Pisa, Florence), Spain, Portugal 4 - 15 October 1582
( Pisa & Florence in 1700)
in Rome, the civil year began 25 December until seventeenth century; in Venice the legal year began 25 March until 1552, &c. &c.
France 9 December - 20 December 1582 in different regions, the year began 25 December or Easter Eve or 25 March until 1564
Netherlands & Belgium: Brabant, Limbourg, Luxembourg, Gelderland (in part), Flanders, Artois, Hainault, Namur, Antwerp, Malines, Holland (including Rotterdam, Amsterdam, Leiden, Delft, Haarlem, the Hague), Zeeland 21 December 1582 - 1 Jan 1583 year began 1 January after 1556, except that:
the year began at Easter in Brabant, Zealand, Holland, Flanders, and 25 March in Delft until 1576
Netherlands: Utrecht 30 November - 11 December 1700 year began 25 December (until 1576?)
England, Scotland, United States 2 - 14 September 1752 English civil and legal year began 25 March until 1752; Scottish until 1600.

Information for other places will be found on Toke Nĝrby's Postal History pages.
For more detailed information, see:
H. Grotefend, Taschenbuch der Zeitrechnung des deutschen Mittelalters und der Neuzeit, (rev. Th. Ulrich), Hannover, 1971.
A. Cappelli, Cronologia, Cronografia e Calendro Perpetuo dal principio dell'Era Crisitiana ai giorni nostri, Milano, 1930.

AH (Anno Hegirae)

The Muslim, or Hijrî, era dates from 15/16 July 622 AD (Julian).  This was the beginning of the Arab year in which the Prophet and his followers made their migration (hijrah) from Mecca to Medina.  Unlike the solar Christian year, it is based on the motions of the moon, 12 lunar months of 29 or 30 days each giving a lunar year of 354 or 355 days in total.
Ideally the beginning of each month occurs when the new moon is actually sighted, and this practice was widely followed for the religiously significant beginning of the fasting month, Ramadan, and the beginning of the following month, Syawal, which marked the breaking of the fast.  A desire for greater convenience and predictability led to the adoption of a civil calendar which did not rely on actual moon sightings, for secular purposes.
The AH and AJ day begins at sunset.

Ru´yah

This calendar is based on months which begin with an attested sighting of the new moon low in the Western sky.  Whether the moon is visible depends particularly upon the longitude, altitude and terrain of the locale, prevailing weather conditions, and the complex interactions of sun and moon. 
The new moon may become visible for the first time when it is as young as 16 hours or as old as 24 hours (approximately).  If the new moon has not been sighted, for whatever reason, by the evening of the 30th day of the month, the new moon is deemed to have appeared, so no month can last longer than 30 days.
The program gives Ru´yah values for Singapore -- which can be used as a proxy for Java and Sumatra with fair confidence -- and for Mekah -- which similarly can be used as a proxy for Cairo, Istambul etc.
Actual local moon sightings have always been subject to the vagaries of weather and human frailty.

For the Muslim legal prescriptions dealing with moon sightings and conflicting dates, see:
Mahyuddin Abu Zakaria Yahya ibn Sharif an-Nawawi, Minhaj Et Talibin  [Minhaj al-Talibin], A manual of Muhammadan Law according to the school of Shafii,  trans. by L.W.C. van den Berg into French and thence by E.C. Howard into English, London, 1914: book 6, sections 1 & 2.
The Ru´yah values used in this program have been extracted from the excellent and well-respected “Moon Calculator” by Monzur Ahmed.  This free DOS program is available for download from Monzur Ahmed's MoonCalc homepage.

al-Battânî & Ulugh Beg

The Muslim civil calendars offer a simple arithmetical appoximation of the lunar calendar.  The recognised civil calendars have certain common features:
  • A fixed pattern in which the 12 months of the year have alternately 30 and 29 days, except that in a leap year the last month has 30 instead of 29 days.
  • A 30-year minor cycle, daur al-saghîr, through which 11 leap (kabîsah) years are distributed. 
  • A 210-year major cycle, daur al-kabîr, comprising 7 minor cycles, after which which the alignment of calendar and weekdays returns to the starting point.
The different versions diverge slightly in the way the leap years are distributed in the minor cycle.  The program gives the two most widely current versions:
  • The calendar of al-Battânî, widely used in the Arab lands, and in the last hundred years in Southeast Asia as well.  This is the version of the calendar generally used by Western scholars.
  • The calendar of Ulugh Beg, widely used in India from Mughal times.
The two calndars differ for one year in thirty, and then only by one day.
Overall the Hisabi method provides an fairly accurate approximation of the ru'yah  calendar, though at any given time there may be a divergence of one, rarely two, days.

The 30-year cycles underlying this calendar begin with 1 AH.  For the principal alternative patterns of kasibah  within the 30-year cycle, see:
Manuel Ocaña Jiménez, Nuevas tablas de conversión de datas Islamicas a Cristianas y viceversa, Madrid 1981, p.31.
Standard tables for manual conversion of Hisabi dates are given in: 
Wüstenfeld-Mahler'sche Vergliechungs-Tabellen zur muslimischen und iranischen Zeitrechnung, various editions.   [al-Battânî]
G.S.P. Freeman-Grenville, The Muslim and Christian calendars, being tables for the conversion of Muslim and Christian dates from the Hijra to the year A.D. 2000, London, 1963.   [al-Battânî]



AJ (Anno Javanensis)

The Javanese calendars are variants of a popular Muslim calendar widely used in the Malay world until about a hundred years ago.   It shares the pattern of alternating 30 and 29-day months with the standard civil calendars, but replaces the thirty-year cycle with an eight-year cycle, known in Javanese as windu. There are three leap years in the eight-year cycle.  This small cycle is a perpetual calendar, very useful for divination based on the Javanese weekday cycles.

Windu, the cycle of years

The first day of each windu  falls on the same weekday.  A simple formula can then provide the weekday on which the subsequent years of the cycle will fall, and then the weekdays on which each month will begin.
The years of the windu  are identified by letters of the Arabic alphabet which have numerical values.  The following example is for a windu  beginning on Friday:
year in
windu
Arabic
letter
Javanese
name
numerical
value
beginning
weekday
1st Alif Alip 1 Friday
2nd (leap) Ha Ehe 5 Tuesday
3rd Jim Jimawal 3 Sunday
4th Za Je 7 Thursday
5th (leap) Dal Dal 4 Monday
6th Ba Be 2 Saturday
7th Wau Wawu 6 Wednesday
8th (leap) Jim Jimakir 3 Sunday
After 15 windu, or 120 years, one day must be skipped to keep this calendar in line with the Hisabi calendar.  When this happens, the windu cycle begins on a new weekday.  Thus, ideally, the 15 "Friday" cycles are succeeded by 15 "Thursday" cycles, and so forth.  The years on which the new series of cycles notionally begin are:
Jumungiah, "Friday" 995 AH 1507 AJ   12 December 1586 AD (Gregorian)
Kamsiah, "Thursday" 1115 AH 1627 AJ   18 May 1703 AD
Arbangiah, "Wednesday" 1235 AH 1747 AJ   21 October 1819 AD
Selasiah, "Tuesday" 1355 AH 1867 AJ   24 March 1936 AD
However the adjustment was not always made at these times, and not at the same time everywhere.  Thus the "Friday" or the "Thursday" style might be found in one place running alongside the "Wednesday" style in another.
The Jumungiah, Kamsiah, Arbangiah and Selasiah calendars began with the styles mentioned above and continued without further adjustment. 

The calendars known to have been in use in the court centres of Solo and (after 1755) Yogya are irregular through neglect and design.  The adjustments introducing both the "Thursday" and "Wednesday" styles were made at different times; and deliberate adjustments were introduced to ensure that the Prophet's birthday (12 Mulud) continued to fall on the weekdays Senen-Pon in Dal years, despite the shift to a new windu  style.  This was considered important because it preserved the Prophet's divinatory horoscope: tradition puts the Prophet's birth in a Dal year.

The pattern used for the court calendars is as follows:
Solo (Surakarta)
Jumungiah 1555 8 July 1633 (Gregorian)
Kamsiah (Dal type 1) 1675 11 December 1749
Arbangiah (Dal type 2) 1749 28 September 1821
Selasiah (Dal type 2) 1867 24 March 1936
Yogyakarta
Kamsiah (Dal type 1) 1681 7 October 1755
Arbangiah (Dal type 1 ?) 1793 6 June 1864
Arbangiah (Dal type 2) 1800 23 March 1871
Selasiah (Dal type 2) 1867 24 March 1936
The court calendars adjusted Dal years thus:
  • type 1:   month 2 (Sapar) is lengthened to 30 days, and month 3 (Mulud) is shortened to 29.
  • type 2:   type 1 changes plus  month 12 (Besar) of the preceding Je year is lengthened to 30 days making Je a leap year, and month 5 (Jumadilawal) is shortened to 29 days making Dal a normal year, though retaining the long final month 12 (Besar) as a legacy of its former quality.

Pawukon, the cycle of weeks

Beside the windu  cycle is the pawukon  cycle of 210 days.  There are 30 weeks (wuku) in this cycle, each named after a legendary guardian, and invested with divinatory qualities.  The pawukon  cycle is also the intersection of the three most significant Javanese weeks, the 7-day week (padinan), the 5-day week (pasaran) and the 6-day week (paringkelan ).  Each week has its own set of weekdays, and any one combination of days in the 7, 5 and 6 day weeks recurs once in every 210-day cycle.  As each of the days belonging to the various weeks and the weeks themselves have a divinatory colouring, the determination of these combinations is the major concern of Javanese calendrical systems.

The 8-year windu  cycles begin with years in which the number of the AJ (or AH) year divided by 8 leaves a remainder of 3.  In each 8-year cycle, the leap years are 2, 5 and 8, except when modified in Dal years. 
The beginning point of each pawukon cycle is the Christian era day number divided by 210 which leaves a remainder of 91.
The Javanese calendars are described in:
G.P. Rouffaer, s.v. "Tijdrekening", Encyclopaedie van Nederlandsch Oost-Indië, vol. 5 (Supplement), pp.401-405, with tables of correspondences between Christian, Hisabi and Javanese dates. Note that Rouffaer assumes that all Dal adjustments are of type 2.  The dates given above for the notional commencement of the Thursday and Wednesday calendars require confirmation by empirical evidence.
A.B. Cohen Stuart, "Nieuwe bijdragen tot de Kennis van de Mohammedaansche Tijdrekening in den Indische Archipel", Tijdschrift van het Bataviaasch Genootschap van Kunsten en Wettenschappen, vol. 20 (1873), pp.195-218.
M.C. Ricklefs, Modern Javanese Historical Tradition, London, 1978.