The following example illustrates the errors modern computer operating systems have with representing historical dates and times.
History tells us that The Great Fire of London occurred on September 2, 1666. We also know that September 2, 1666 was a Sunday. Here is a banner from The London Gazette newspaper published at that time.
This particular newspaper, a weekly, published Monday to Monday. Monday was September 3. Sunday would have been September 2.
Yet a modern computer program will insist that September 2, 1666 is actually a Thursday.
| |
.Net Calculation | Thursday |
| |
Java Calculation | Thursday |
|
The reason for the error is because modern calendar programs (those tied to the OS) use the Proleptic Gregorian Calendar to determine any date. So, we need to talk about that.
The current calendar that most of the world uses today is the Gregorian calendar, but as you regress in time, changes to the
calendar were made. Unfortunately, they not made in a uniform manner. The changes to the Gregorian calendar were largely
based on location and time. Here are two classic examples:
October 1582, Italy
CENTER> |
Sun
|
Mon
|
Tues
|
Wed
|
Thur
|
Fri
|
Sat
|
|
1 |
2 |
3 |
4 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
31 |
|
|
|
|
|
|
|
September 1752, England |
Sun
|
Mon
|
Tues
|
Wed
|
Thur
|
Fri
|
Sat
|
|
|
1 |
2 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
|
As you can see from each of these calendars, dates were removed. We talk a lot about the removal of calendar dates in other articles, but suffice it to say, the removal of the calendar dates signifies a moment in time when the location in question transitioned from the Julian Calendar to the Gregorian. In 1582, October, it required only the removal of 10 days from the Julian calendar to transition it to the Gregorian; by 1752 in England, it required 11.
That is because as you advance in time, the further out of sync the Julian calendar was becoming with the seasons. Its year length and its year length rules actually created a calendar that was longer than the year it was describing. The Julian Calendar has 365 days a year and a leap year every 4 years, or 100 leap years every 400 years. The Gregorian calendar has 365 days a year and 97 leap years every 400 years. Anyway, back to 1582 ...
The Gregorian Calendar in 1582 underwent a massive change but only for a handful of countries. Countries like
England, for example, didn't follow suit until 1752 ( after the Great Fire of London ) and their reformed calendar looked quite different. So if you go back in
time, would you have found an October 5, 1582? If you lived in Italy, no. If you lived in England, yes. If you lived in
England, Scotland, Sweden, yes, there would have been an October 5, 1582, as England, Scotland, the American colonies didn't
convert to the Gregorian calendar until 1753. Sweden was the year later.
So what did the calendars look like for the year 1666? Side-by-side we show you September 1666 calendar from the perspective of a native Italian and the other from the perspective of a Londoner.
September 1666, Italy (Gregorian Calendar)
CENTER> |
Sun
|
Mon
|
Tues
|
Wed
|
Thur
|
Fri
|
Sat
|
|
|
|
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
|
|
|
September 1666, England (Julian Calendar) |
Sun
|
Mon
|
Tues
|
Wed
|
Thur
|
Fri
|
Sat
|
|
|
|
|
|
|
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 |
28 |
29 |
30 |
|
|
|
|
|
|
|
The Proleptic calendar is an attempt to effectively deny reality. It begins at 1/1/1 and applies Gregorian calendar rules
to pre-Gregorian dates. The earliest date on the Gregorian Calendar is October 15, 1582, but only for Italy and a few other countries. Yet the Proleptic Calendar would give you the impression that the Gregorian calendar has been used since 1/1/1. It ignores the concept of calendar reform (of skipped or missing days) to result in a 'average'
calendar across this time span.
The whole function of a calendar is to present with accuracy a past date, and display with accuracy a
future date. Anyone can display a past or future date incorrectly. That takes no skill.
Prior to the adoption of the Gregorian calendar, the leap year rules were very simple. Every 4 years. With the adoption
of the Gregorian calendar the rules became more complicated: Every 4 years, unless the year is divisible by 100 then there is
no leap year, unless the year is divisible by 400 then there is. Essentially, the Gregorian calendar has 97 leap years for every 400 years; the Julian has 100 leap years for the same time period. The Proleptic calendar is an attempt to apply Gregorian
calendar rules to non-Gregorian (Julian) dates. The result is quite a muddle. As you can see from the table below,
historically, there was a February 29, 100, as there was a February 29, 200 and so on. Yet the Proleptic
calendar dismisses them out of hand as days that simply don't exist.
That's rather inconvenient for the historian who is working on old shire records or scrolls that happen to mention these
dates.
Date |
Julian Calendar |
Proleptic Calendar |
Gregorian Calendar |
February 29, 100 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 200 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 300 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 500 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 600 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 700 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 900 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 1000 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 1100 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 1300 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 1400 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
February 29, 1500 |
Existed - was a leap day |
Date does not exist on calendar |
Calendar did not exist |
October 5, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 6, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 7, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 8, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 9, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 10, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 11, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 12, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 13, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 14, 1582 (Italy) |
Existed |
Existed |
Calendar did not exist |
October 15, 1582 (Italy) |
Existed |
Existed |
Existed |
For us, it simply comes to a matter of trust. How can you trust the Proleptic calendar to be accurate with a future date,
if it is to widely inaccurate about past date representation?
Our philosophy at BlackCat is very simple: do it right. So much more is at stake than just the extra or missing calendar days: phases of the moon, sun rise and sun sets, in short the very rhythm of the heavens. If you were researching the life of England's King Richard III, and needed to know what was the phase of the moon the night before the Battle of Bosworth in August 22, 1482, which calendar, which astronomical calculations would guarantee you a correct result?
The calendar that is NOT the Proleptic one would. A Proleptic calendar in this context is nothing more than a wild guess.
|