Talk:Julian and Gregorian Calendars

From FamilySearch Wiki
Jump to navigation Jump to search

When the Gregorian Calendar began, there were many people who wanted to go back and reinterpret all the previous dates into the new Gregorian Calendar system. You can see this even today in several websites about the Gregorian Calendar. This would remove any doubt due to the change over to the new calendar. Also, it would remove any lack of standartization, since the Gregorian Calendar did not start at the same time in every country. This is kind of like rewritting history to accomodate a new idea. 

But think what this would entail. Millions upon millions of old dates would have to be changed. Also, how would you ever document an old date. If the source gave a date in the Julian Calendar that now had to be reinterpreted into the Gregorian Calendar to a different date, can the source be used any more as the documentation for this new date?

For these problems and maybe others, I propose the following. If the date falls before Pope Gregory gives his decree on the Calendar in 1582, then the date is assumed as following the old Julian Calendar. If the date falls after Pope Gregory gives his decree on the Calendar and before it is adopted in a particular country, then write it as a double-year date. Ignore any days that might be skipped. For example 2 March 1699\1700. At the time it would be considered 1699 following the Julian Calendar, then in effect. But now it would be considered 1700 following the Gregorian Calendar now in effect. Finally, if the date falls after the Gregorian Calendar is adopted in that country, then the date is assumed as following the Gregorian Calendar.

To me this is the only reasonable way to handle all dates, those following the old Julian Calendar and those following the Gregorian Calendar. Please add comments. Sabwoo 02:19, 5 October 2012 (UTC)


It is unfortunate that Britain decided to change the start of the year from 25th March to 1st January at the same time as changing from the Julian to the Gregorian calendar because the two changes have become conflated, as in the contribtion above. 

There are  two completely separate issues. Pope Gregory made no change at all in the start of the year, he just continued using 1st January as in the Julian calendar introduced by Julius Caesar in 45bc and as used in Rome before that. 

But many countries used the Julian calendar for the length of a year but adopted their own start date for a year. Britain adopted 25th March in 1155ad and changed to 1st January in 1752 so double dating should apply from 1155 to 1752.  I suspect that many genealogists do not realise that a date written in January 1580 should be shown, logically, as 1580/81 just as much as one in 1585 should be 1585/86. It has nothing to do with whether a Julian or Gregorian calendar was in use.

This issue is particularly important when handling dates from 1538 when parish registers started in Britain.  It might also apply to wills and manorial, court and other records before 1538 but most of these used regnal dating where this issue does not arise. cbeobe 12 November 2014

Which calendar?[edit source]

Julian and Gregorian are just two of many calendars. Clearly, to move forward as one tree with global historical records and global contributors, Family Tree will need more standardization. Toward that goal, Family Tree may need a date menu similar to its new and increasingly helpful place menu. The date menu would offer an appropriate selection of calendars and be further localized as to language and script.

The way to prepare for this is to observe the Family Tree foundation principle of respecting historical records. Record the date as it appears in the historical record, without attempting conversions to adjust for later changes.

Dontiknowyou (talk) 15:28, 8 August 2020 (MDT)

You raise a valid point here. However, the FamilySearch Research Wiki is not the appropriate venue for a discussion on how FamilySearch's Tree application deals with differing calendar systems and implements a solution. I suggest you contact the team responsible for the Tree directly to communicate your concerns. SvareJM (talk) 14:45, 11 August 2020 (MDT)
This talk page is the appropriate venue to discuss advice given to contributors on this page. I think the current advice is flawed. That advice is To avoid any confusion, write the date with both years' numbers. For example - 14 February 1699/1700. At the time it would have been considered 1699 according to the Julian Calendar, then in effect. But now it would be considered 1700 according to the Gregorian Calendar. Using the double-year dating and understanding its purpose can be helpful in recording historical events. Is it appropriate to give any advice? Perhaps the quoted text should just be deleted? It actually creates confusion. Dontiknowyou (talk) 10:00, 1 September 2020 (MDT)
I have shared and discussed your concerns with other wiki staff and FamilySearch's Chief Genealogical Officer. We have decided that as there is a need to preserve the genealogical soundness of published and written records which employ double dating (or dual dating) as well as inform new researchers of the practice so they can correctly evaluate information they may be presented, we will keep the advice as it is presently written. The practice itself is well-documented, and examples can be found dating back to the calendar change itself. There has been a significant amount of discussion, both scholarly and informal, on the topic and the recommendation is always to preserve the double-dating. I would recommend to you Mike Spathaky's article Old Style and New Style Dates and the change to the Gregorian Calendar: A summary for genealogists for your reference. SvareJM (talk) 18:17, 1 September 2020 (MDT)
I am glad to learn there has been discussion, albeit behind the scenes. It seems we all agree dates should be kept as they appear in historical records. Yes? If that is the case, this instruction is not clear. It can be read as instructing contributors to "correct" dates on Family Tree profiles. See for example the birth date of Samuel Clough LZF4-BN6, and its number of contributors, comments in the change log, and notes. --- Dontiknowyou (talk) 17:13, 2 September 2020 (MDT)
Yes, we are in agreement that dates should be kept as they appear in the records, preferably, the original record and not a transcript. Your additional comments go back to my original point regarding how Family Tree deals with dates. The engineers have had the issue raised to them several times, and if, or when an update to the system to accommodate dual dates is planned has not been shared with Research Wiki staff. I recommend you use the Feedback link at the bottom of the page on Family Tree to share your concerns with them. SvareJM (talk) 09:57, 3 September 2020 (MDT)

When does the year begin? (revised)[edit source]

As noted above, when the calendar year begins is a separate issue from which calendar is used. In Britain this issue is historically very complex. Garmonsway's introduction to The Anglo-Saxon Chronicle reviews the complexity and gives a summary (Garmonsway 1953, page xxix):

Consequently there is the possibility that the Chronicle may have used any of the following dates, set out in chronological order, for the opening of the year:

(a) 25 March preceding our 1 January: the Annunciation (Stylus Pisanus).
(b) 1 September: the Greek or Byzantine Indiction date.
(c) 24 September: the Caesarean Indiction date (Mid-Autumn Day).
(d) 25 December: Christmas Day (Mid-Winter's Day).
(e) 1 January.
(f) 25 March following our 1 January: the Annunciation (Stylus Florentinus).

This FamilySearch wiki page (meaning article, not its talk page) now says:

To avoid any confusion, write the date with both years' numbers. For example - 14 February 1699/1700. At the time it would have been considered 1699 according to the Julian Calendar, then in effect. But now it would be considered 1700 according to the Gregorian Calendar. Using the double-year dating and understanding its purpose can be helpful in recording historical events.

Reasons to abandon this instruction include:

  1. The instruction conflates a change of calendar year opening, from 25 March to 1 January, with a change of calendar from Julian to Gregorian.
  2. The instruction is appropriate only for a subset of historical dates: those recorded within certain limited, ill-defined geographic, political, and temporal bounds of Britain and British Colonial America.
  3. Most Family Tree contributors neither care nor know which calendar or calendar year was in use when historical record was made.
  4. Instructions that are difficult to follow should be avoided, because they produce inconsistent results.
  5. Family Tree does not standardize dual dates nor should dual dates be standardized, for reasons above.
  6. Non-standardized dates cause Family Tree hints that are in error and confuse / frustrate contributors.
  7. (new) Per discussion above, the intent of the instruction is to preserve dual year dates found in historical records, not to instruct contributors to calculate "correct" dates.

--- Dontiknowyou (talk) 10:22, 4 September 2020 (MDT)

Use case: Samuel Clough[edit source]

The use case that brought me to this page is the birth date of Samuel Clough LZF4-BN6:

Samuel Clough birth date.png

The historical record gives his birth date as 20/12/1656. Written numerically, the date has both Julian/Gregorian calendar year and year opening issues. In 1656 in British Colonial America the Julian calendar was in use and, furthermore, the 12th month of the year was February, not December.

Note the very large number of contributors, reflecting the expenditure of much time, effort, and frustration.

The birth date on the profile is now delicately arranged so as to not trigger a "Missing Standardized Birth Date" alert and has an explanatory comment. But this is all too much trouble!

--- Dontiknowyou (talk) 10:22, 4 September 2020 (MDT)

Proposal[edit source]

I propose the following new, simple instruction:

  1. Standardize the date as in the historical record. If there is a problem with the date use the comment field to alert other contributors.
  2. If the historical record gives a dual date (such as 14 February 1699/1700), record it as such without correcting it and explain in the comment field.

Here are some examples of problems and how to solve them:

A birth on 20 December 1699 and death on 14 February 1699 for an infant in British Colonial America. In British Colonial America at that time the calendar year began on 25 March, so despite appearances these records are in the correct order: birth before death. Family Tree does not adjust for changes in calendars and year openings, and neither should you. If you know the date was recorded when and where the year did not open 1 January, add a brief explanatory comment and dismiss the "Birth Before Death" alert.
A birth on 12/20/1656. Standardize the date as 20 December 1656. If you know the date was written in a place and time where the 12th month was February, standardize the date as 20 February 1656. Do not apply the dual date convention. If you know the specific adjustment needed, put that information in the comment.

--- Dontiknowyou (talk) 10:22, 4 September 2020 (MDT)