Avram Friedman wrote
| If I may Measuring time is not science or math based, it is observation based.
followed by a recapitulation of some of the features of the Hebrew religious
calendar. (This calendar is treated systematically in chapter 9 of Dershowitz
and Reingold, Calendrical calculations.)
Now science is observation-based, but what Mr Friedman has in mind is naif,
local , naked-eye observation.
This tension between a promulgated, official calendar and local observation is
a very old one. In the United States different local times, fixed with
attention to local latitude and longitude coordinates, were the rule in
essentially all communities until the middle of the nineteenth century, when
the coming of the telegraph and railroads brought time zones and inexorable
standardization.
The Hebrew calendar is a lunisolar one, and like all of them it attempts to
reconcile the irreconcilable: the orbit of the earth about the sun and that of
the moon about the earth are not simply related.
The Islamic calendar, on the other hand, is an all but pure lunar one, and
there is a Koranic injunction against the use of leap months like those
inserted inoto the gravid or embolismic years of the Hebrew calendar. (The
Islamic calendar does have leap days of a sort, but that is another matter.)
In Moslem countries a civil calendar, typically--except for Iran and
Afghanistan--the Gregorian calendar, is used for most purposes, with the
Islamic calendar reserved for ritual, religious purposes. Ramadan (Ramazam),
for example, begins and ends when the relevant new moons are actually observed
locally. This exclusively lunar scheme---An Islamic-calendar year contains a
mean of only 354.36666... days as opposed to the mean of 365.2425 days in a
Gregorian year---is not synchronized with the seasons. This precession would
perhaps be a fatal disadvantage if this calendar were used for secular
purposes, but for religious purposes it is perceived to be an advantage: the
fasting rigors of Ramadan fall sometimes in winter and sometimes in summer, and
they are not uniformly more onerous in some parts of the world than they are
elsewhere.
For secular purposes, however, a precise lunisolar calendar is a necessity, and
the science of the National Observatories underlies it. Naif local observation
cannot and does not play any role in the maintenance of the modern Gregorian
calendar.
One of the perhaps obvious advantages of the Day Serial Numbers (DSNs) that I
described in earlier posts is just that they make date conversions between two
calendars, even two such different ones as the Gregorian and Islamic calendars,
trivial: Day 1 of the Islamic calendar is---It began at sundown on
Thursday---Friday, July 16th, 622 CE of the Julian calendar. This Julian
calendar date has the Gregorian DSN 227015. Et voilà! The conversion is in
effect accomplished.
It would be disingenuous not to concede the religious and the scientific temper
are often in sharp conflict, but no religious observance need be compromised in
any way by the use of a scientific, secular calendar for secular purposes.
Religious observances are even facilitated by the availability of
secular-calendar estimates of the most probable dates of events like Ramadan
and the Islamic New Year (Muharram 1).
What is needed is some considerable understanding of the structural differences
among calendars: A perhaps not quite obvious consequence of the relative
shortness of the Islamic-calendar year is that annual Islamic holidays can
occur twice in a Gregorian year; or again certain Jewish holidays---minor,
non-seasonal ones fortunately---do not occur at all in some few Gregorian
years.
Anomalies of this sort can be disconcerting to the uninformed. More important,
they impose design requirements on programs. Calendrical routines that
determine the Gregorian dates of other-calendar holidays must in general be
prepared not for a result that is a single scalar date|DSN but for one that is
a vector of zero, one, or two dates|DSNs.
John Gilmore Ashland, MA 01721-1817 USA
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
