If I have a date on a birth certificate that says: "January 13, 1719", do I write the year as 1719, 1718/19, or 1719/20 (the latter two being double dates).

What if the date is April 28, 1719?


Okay. I obviously didn't make this question clear enough. Let me try to make my question clearer:

Genealogists use double dates to try to prevent misinterpreting a Gregorian date for a Julian date and vice versa.

Let's say I have a record containing a date from this confusing time period.

  1. How do I know whether its a Julian date or a Gregorian date?
  2. If it's a Julian date, how do I display it?
  3. If it's a Gregorian date, how do I display it?
  • I've not run into double-dates. Wouldn't a person just have a single day they were born on?
    – Justin808
    Commented Oct 10, 2012 at 5:29
  • 3
    @Justin808 and others who don't know what double dates are: See genealogy.com/00000771.html
    – lkessler
    Commented Oct 10, 2012 at 5:42
  • 1
    @lkessler You might add a summary of that linked rationale to your questions ... least folks thinks we are double dating in the genealogy & family history Q&A.
    – GeneJ
    Commented Oct 10, 2012 at 5:55
  • 3
    Hey! This is supposed to be a site where expert genealogists answer questions. If you don't know what a double date is, then don't answer. Seriously though, I've tagged double-date and proposed a wiki definition for the tag (which still needs to be approved by some moderator). Shouldn't that be good enough to allow someone who is not experienced to find out what it is?
    – lkessler
    Commented Oct 10, 2012 at 6:02
  • 1
    @lkessler If you are looking for an exercise in double-date interpretation examine Example 1 in genealogy.stackexchange.com/a/1551. What was the timespan in days between the two deaths (and presumed births)?
    – Fortiter
    Commented Oct 19, 2012 at 2:24

7 Answers 7


As you have learned, your question(s) is more complicated than it first seems.

You can not tell if a single date is Julian or Gregorian unless something else in the record provides additional information. If the date is written with two years, then you have the information that the author recognised the potential confusion.

In some cases, the suffixes OS (old style) or NS (new style) may be added to a single year in the same manner as we can use BCE and CE. So 16 February 1583 N.S. is an explicit acknowledgement that the date is recorded in terms of Pope Gregory's edict (although the original abbreviation would probably be in Latin!) https://en.wikipedia.org/wiki/Old_Style_and_New_Style_dates provides some examples of this and mentions similar forms in other languages.

If you are certain that a given date is one or the other, you could add a note to the effect that you know this "date is OS because ...".

If you are uncertain which calendar has been used, then you should record that as a note (and seek further information). Trying to guess which system has been used can only introduce the possibility of more error.

Note also that there are circumstances where more than the year is duplicated. Contracts between merchants in England and Russia in the nineteenth century needed to specify different dd and mm even when the yyyy portion of the date was the same.

  • I think the old style / new style designation that you refer to is the key when the info is there, and I do admit that I've seen this designation in some of my ancestors' Romanian certificates without knowing at the time what that meant. Your wikipedia link really gives a lot of information that I didn't know before and it leads to this great tool: 5ko.free.fr/en/jul.php
    – lkessler
    Commented Oct 14, 2012 at 20:36

Would you not record what was on the birth certificate?

  • I second this idea. Original date and place of issue of a document give enough information to uniquely identify the day. The software should know all the time changes. This way you keep your records correct also on the international level, which is important in genealogy, see answer by Gene.
    – texnic
    Commented Oct 10, 2012 at 7:29
  • 4
    No. You record what was on the birth certificate in the source information. You record your interpretation of what the date was in your conclusion about the event. I want to know what the interpretation should be.
    – lkessler
    Commented Oct 11, 2012 at 3:38
  • As part of my citation, I would record or "quote" the date exactly as written in the record. When I entered the information in the tag/even/pfact, I would assess and reflect the double-date.
    – GeneJ
    Commented Oct 16, 2012 at 20:38

Assuming that "January 13, 1719" is in the range where double date notation is appropriate, which will depends on when the Julian to Gregorian transition took place in the area where the source was generated, then what you're really asking here is "how do I know what the author of this document meant?".

The answer to that of course, is that it is not possible to know for sure. The author may have been assuming a year start of January 1st, or of March 25th. If you can find additional sources for the date in question then that may make it easier to come to a conclusion as to whether "1718/9" or "1719/20" is appropriate as the year but otherwise you are probably left trying to work out which is more likely.

The "April 28, 1719" date however is not ambiguous, as it falls after Lady Day (25th March) and hence was in 1719 in both calendars, so never needs to be written as a double date.


It depends on the document source and how the application you are using deals with dates.

For the January example with no other information I would record that as 1718/1719. The question needs to be asked though how is the source document written. Are you accurately transcribing a source record or using it for some other purpose.

The April example is much easier to answer you would record that as 1719 as by April the two calendars were in sync.


Short answer: Write 1719

Under the Julian/Gregorian calendar system there has always been a lot of opinions on when the year starts, including having liturgical years that doesn't align with the common years. This means that in the time there is overlap it's unclear which year is meant.

For example:

Under the Julian calendar in most of Catholic Europe the first day of the year was March 25, but with the Gregorian calendar, the Catholic church moved the first day of the year to January the first (and the lithurgical start of the year to first Advent, but let's ignore that for now). Since England and several other northern European countries had broken from the Catholic church, they kept the Julian calendar for a long time. Therefore, when they wrote "January 13, 1719", the Catholic world had already switched to the year 1720. But the practice of having January 1st as new year crept into England as well.

This means that any date that falls in between January 1st and March 25 can be in one of two years, depending on if it's written with the new new year or the old new year in mind.

To solve this, they introduced double dates, to specify clearly which year it is by writing "January 13, 1719/1720", to clarify that this is 1719 old style and 1720 new style.

So, what should you write when your source says January 13, 1719? You should write 1719. The reason for this is that except with extra proof you don't know if 1718/1719 or 1719/1720 is meant. The double dates are written to specify the year exactly, and since your source didn't, then neither should you.

But what if you know by other data that 1719/1720 is meant? Well, even then I'm not sure you should use the double date format, as it's confusing. People in general tend to understand that it means "approximately", and how can something be at January 13, but only an approximate year? You only understand what double dates mean when you know about these issues. So it doesn't really clarify anything to the uninitiated, while those who know about this do not need help clarifying it.

It also hides and ignores the fact that although this is January 13, 1719 in the Julian Calendar, it is not January 13, 1720 in the Gregorian calendar. It is in fact January 24, 1720 in the Gregorian calendar.

So, unless restricted by software, which rarely supports these kind of things, I would instead mark the date as being in the Julian calendar. Depending on context you can also add more information, like a footnote. Or make a proper conversion and write both dates, if that's meaningful.

However, of course, if the source is using a double date, so should you. Preferably with a note explaining what's happening. ;-)


Depending on the software you are using (unless you're doing it all by hand) you may be able to enter both dates separately choosing the type of date being entered, or by the software configuration pick a default to use.

If the software doesn't support it you could always use one format or the other and simply convert the date before entering it. This way you know all dates in your system are in a single format.


Dates are quite a bug-bear to the genealogist. I've had problems in the past exchanging data with Colonial Cousins who insist that a date written D-M-Y is in fact M-D-Y...

That is a 'problem' only because the USA insists on using a different format from the rest of the world. Also, the genealogical standard is to write D-M-Y

But for double-year dates, I tend to use the 1718 date if I'm forced to use just one

As an aside, for your example of April 28, 1719 then this is outside the double date window, so will be as written.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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