For dialogues occurring in the story the characters would use their respective culture/civilization calendar, but for the one writing the story a way to use a familiar calendar to place an event is needed sometimes.

So, I usually use an spreadsheet to place events, later I started to write my own graphical tool to place events in an intuitive timeline. I must say that fantasy calendars for fantasy worlds are far easier to implement that our current calendar. Let's not speak about others calendars like the original Japanese calendar.

Everything is fine as long as you don't need to overlay a calendar over another, for example if the stories doesn't need to touch the topic of dates or you can change things to avoid it. Problems start when you actually need to figure out dates in more than one calendar.

I have the formula to correctly calculate leap years, but something confuses me about those 10 dropped days when they synced the Julian calendar with the Gregorian calendar.

Those 10 days are to sync the two calendars, correct? It doesn't mean that the Gregorian Calendar has a hole of 10 days, correct? But if I were to work only with the Gregorian one, do I need to cut out those 10 days? Or said in another way, does 0001-01-01 in both calendars refers to the same date or one is 10 days before the other?

  • 3
    $\begingroup$ Close voters: You should explain why you are voting to close. This is a technical question about timekeeping related to Worldbuilding. How is it 'story based?' $\endgroup$ – kingledion Mar 23 '17 at 18:55
  • 2
    $\begingroup$ I'm asking the close voters, I think your question is fine. $\endgroup$ – kingledion Mar 23 '17 at 19:09
  • 7
    $\begingroup$ I think some of the issue is that this is a question that has more to do with a real life concept, and theres a general preference to put those on other Stack Exchanges. Personally, I think that the particular topic of calendars has enough to do with WorldBuilding that it belongs here, and I'm not entirely sure which other StackExchange it could go on! They might also be closing because the information you ask for is technically on Wikipedia, but calendars are so complicated that I think asking for clarification beyond Wikipedia is totally fair game! $\endgroup$ – Cort Ammon Mar 23 '17 at 19:47
  • 1
    $\begingroup$ @CortAmmon I must admit, that I was a bit worried that this may not go in worldbuilding. But in the end, I concluded that real life calendars are relevant for worldbuilding. $\endgroup$ – Hatoru Hansou Mar 23 '17 at 19:52
  • 4
    $\begingroup$ That and, after researching the answer myself and then finding out how many corrections I needed to make... I'm starting to think our real life calendars are more fantastical than any fiction I've read! Who were these people, and why were they not locked up?!? $\endgroup$ – Cort Ammon Mar 23 '17 at 22:36

Dates between the years AD 201 and AD 299 in both the Julian and the proleptic Gregorian calendars are the same for the same day. Before 201 Julian dates are a little "later" than the proleptic Gregorian dates for the same day; for a random example 14 August 60 AD Julian is the same as 12 August 60 proleptic Gregorian. After 299 Julian dates are "earlier" than the proleptic Gregorian dates for the same day; for a random example, 14 August 333 AD Julian is the same day as 15 August 333 proleptic Gregorian.

To your specific interest in days near the start of the common era, The Julian January 3 1CE corresponds to the Gregorian January 1 1CE, according to this calculator.

The issue solved by the Gregorian calendar was that the Julian year was a hair longer than the tropical year (aka solar year - the time it takes for the sun to return to the same position, such as vernal equinox to vernal equinox). Over time, the errors had accumulated such that the Julian calendar date where any given yearly celestial event occurred had drifted by 10 days. In other words, even though the events were supposed to be yearly, the events were occuring at earlier dates every year. This was causing trouble for the Church which tied Easter to the vernal equinox but which had also pegged the equinox to Mar 21. During the era when the Gregorian calendar was developed, the equinox was instead occurring for the Church around Mar 11.

By having the Gregorian calendar start at the same time-point as the Julian calendar, but with a leap year algorithm fine tuned to be more in line with the celestial bodies, the Gregorian calendar moved the equinox back to Mar 21 (an ecclesiastically set date from back in the 3rd century, before the drift had become noticeable).

Different algorithms were suggested. One suggestion was to remove the leap years for the next 40 years. This would have caused no calendar date jump (no "days lost"), but would have lead to 40 years of special cases written into the algorithm forever. Instead, they chose to rip the bandaid off quickly. They decided to instead declare that they would switch from the Julian calendar to the Gregorian calendar (which both start on the same day but have different leap year algorithms) cold turkey:

When the new calendar was put in use, the error accumulated in the 13 centuries since the Council of Nicaea was corrected by a deletion of 10 days. The Julian calendar day Thursday, 4 October 1582 was followed by the first day of the Gregorian calendar, Friday, 15 October 1582 (the cycle of weekdays was not affected).

So, if you are using the Gregorian calendar, dates before 1582 will continue to follow the same pattern as the rest of the Gregorian calendar. If you think about it as though you are going back in time, the leap years will slowly cause the two calendars to sync until they are perfectly synced between 201 AD to 299 AD.

  • 4
    $\begingroup$ For those who think this is interesting and want to dig further, a recommendation I have: whenever you come across a calendar discrepancy like this, find three different phrasings to describe what is going on, ideally from three different perspectives. I have found (empirically) that that third point of view is incredibly helpful for preventing me from misinterpreting the results. In this case, the viewpoints are a) The monks in 1582 trying to figure out how to make things work, b) The monks in the 3rd century fixing Easter, c) the motion of the celestial bodies when viewed from earth. $\endgroup$ – Cort Ammon Mar 23 '17 at 18:55
  • 2
    $\begingroup$ I had a hard time getting myself to continue past "Over time, the errors had accumulated such that the Julian calendar date where celestial events had drifted by 10 days." Did you mean "... where celestial events line up with the Gregorian calendar had drifted by 10 days"? (or something similar?) $\endgroup$ – Loduwijk Mar 23 '17 at 21:13
  • 3
    $\begingroup$ @AlexP Oof! Even more complicated than what I thought when I wrote it! I've edited the answer slightly to fit with what you point out. Man, I'll take TAI vs. UTC any day over Gregorian vs Julian. This calendar stuff is just as brutal as I remembered it was! =) $\endgroup$ – Cort Ammon Mar 23 '17 at 22:10
  • 2
    $\begingroup$ @CortAmmon actually I think it is always more brutal than I thought it was no matter how bad I thought it was.... :) $\endgroup$ – Erik Mar 24 '17 at 2:12
  • 1
    $\begingroup$ For added complication, although Italy, Spain, (most of) France and others adopted the Gregorian calendar in 1582, it took more than 300 years for the rest of the world to catch up (Turkey not switching until 1927, when 13 days were skipped) -- see Change From Julian to Gregorian Calendar. $\endgroup$ – TripeHound Mar 24 '17 at 11:48

Simply put, neither Julian nor Gregorian calendars have holes in their own dates. They have a point in time that is the same date for both. After that, the 2 calendars are in sync for some time, but later start to diverge. For any single day after the divergence, 2 different dates exist, one Julian and one Gregorian.

Since only one of the dates is normally used in everyday life, switching between the calendars requires to change the current date non-continuously on the switch day, creating the gap in the numbering. The gap was 10 day initially, but some countries joined lately, and their gap is larger (13 days in Russia, for instance).

On the switch, there was never any intent to synchronize the calendars at the switch date. The point was to synchronize the spring equinox to its traditional location on the calendar, stopping its drift accumulated over a millenia. So the gap was expected and allowed.

For some time after the switch, people tended to use dates from both calendars simultaneously for same days (sometimes called "old style date" and "new style date").

Although the Gregorian dates weren't really used before its establishment, the calendar can be extrapolated back in time; it's called proleptic calendar. With this system, most days in history have dual dates, but in practice it only confuses everyone, so it's not normally used. Please note that the divergence between Julian and proleptic Gregorian is not actually at the year 1, so 0001-01-01 dates in the 2 calendars don't match.

So the best way to date a day before 1582 would be to just use its Julian date. That's what people actually used then, and that's what was written in any historical document.

  • $\begingroup$ Actually, after reading about proleptic calendar (thank you), any in-story reference to dates would be Julian. But I would want to place the event in a proleptic calendar timeline. $\endgroup$ – Hatoru Hansou Mar 23 '17 at 19:21
  • 6
    $\begingroup$ For added spice, remember that dates are really time and place dependent. For example, before 1752 in England years began on March 25th, so for example 24 March 1648 was followed by 25 March 1659; up to 1600 or thereabouts, Greeks, Russians and other Orthodox nations used the Byzantine era, which is like the Julian calendar except that years began of September 1st and were counted "since the Creation of the world", so that for example 1 September 1400 "after the birth of the Savior" was dated 1 September 6909 "since the Creation". $\endgroup$ – AlexP Mar 23 '17 at 22:15

Time is complicated and difficult. Like you said it is best just to hand wave dates if you can. For example you mentioned 1582 as the adoption date but that is very dependent on your story's setting. For example France was 1582, but Prussia was 1610, and England was 1752 (and only had 282 days in 1751 as a consequence). As the English example shows years didn't even turn over at the same time everywhere.

If you can't hand wave and have some proficiency programming I suggest you save yourself some hassle and write a simple program to do the conversions for you. If you know Java then the wildly popular Joda Time library is a good fit, and as you can see by the link actually attempts to support this transition. If you prefer .NET then Noda Time is an attempt to bring similarly improved date/time handling to .NET. This still doesn't make everything perfect though because time is hard, and countries in the past and present are able to adjust things by fiat. One example is the "New Years Day" isn't uniformly January 1st under the Julian calendar.

So my advice is just try to get things reasonably close to what you want in a way that will make sense to your readers. Don't worry too much about getting it wrong because you're almost certainly are going to get some things wrong. If the natives of the age didn't care enough to have a sensible standard then why should you beat yourself up about minor errors? You can even argue that errors are more realistic since it was easy to make these mistakes during this time period.

  • $\begingroup$ Libraries are welcome but as I need to work with fantasy calendars too I may very well implement everything myself. Yes, for most cases you can avoid the topic of calendars, in probably 99% of the cases nobody cares about Julian/Gregorian/others. At production, when planning things, most of the time the only thing that matters are the order of events (maybe their duration). But then let's suppose the theme is "time travel", I can think in at least one situation where you can have some fun with the calendars. $\endgroup$ – Hatoru Hansou Mar 24 '17 at 0:08
  • 1
    $\begingroup$ @HatoruHansou I see what you're shooting for now. Of course if you have a time machine you would need to have some sort of "universal" calendar that is more orderly than any calendar known to man which would be your fantasy calendar. Unless you've already solved that problem I would think that a question on the basis of a time travel calendar would be even more interesting than your current question. $\endgroup$ – Erik Mar 24 '17 at 2:10

Not the answer you're looking for? Browse other questions tagged or ask your own question.