Dates in databases

Back to Hints & Suggestions

About the Database contains a section on ‘Dates’ which gives a taste of my frustration when dealing with dates in genealogical databases. Too often I have found that different sources gave different birth, marriage or death dates for people we were investigating. Presumably, many of these discrepancies came from simple error (mistaken identity, wrong reading or wrong translation of writing which was difficult to read or understand, etc.) or sloppy work (typographical slips, being unable to read your own notes, etc.) but there are sometimes good grounds for confusion.

 

Here are three issues relevant to British & Irish families which are worth drawing to your attention.

1. For many centuries, whilst people frequently identified years by reference to ‘Anno Domine’, most legal documents were dated by reference to the reign of the Monarch; eg. 1EdwII (or 1EdII) signified the 1st year of the reign of King Edward II. His father died on 8th July 1307 so, even though he was not crowned until 23rd February 1308, it is generally accepted that he became King on that date. Hence, 1EdwII lasted from (some hour in) 8th July 1307 to (that hour in) 8th July 1308. Hence, for example, 3EdwII would have included August 1309 and May 1310. Whilst September 3EdwII would therefore mean September 1309, over time people forgot (or did not know and could not be bothered to check) exactly when the reign started so would either make an assumption (often just adding 3 to the year when the monarch came to the thrown so that September 3EdwII would be shown wrongly as September 1310) or would show two years to cover the position (eg. September 3EdwII recorded as September 1309-10).

2. Whilst calendars may now be viewed as fixed, with days & months ‘set in stone’ so that (for example) 20th June 1981 means 20th June 1981 and is unlikely to be confused with any other date, that was not always the case. There was no such date as 10th October 1582 in Spain even though there was in England. There was no such date as 10th September 1752 in England (and the rest of Great Britain) but there was in Spain. In both those cases, the date came just after the relevant country had changed its calendar from the Julian version, which had existed since Roman times, to the Gregorian version which we have today. The first note below explains the change.

3. When the Julian Calendar was first adopted, 1st January was set as the first day of the year. However, the influence of the Christian Church meant that some countries came to view certain days that had religious significance as the first day. In England & Scotland, for example, even though 1st January continued to be celebrated as a festival, in medieval times 25th March (Annunciation or Lady Day, being 9 months before the birth of Christ), became identified as the first day of the civil calendar. More on this in the second note below.

 

A brief (simplified) history of the calendar in Europe

The story of the development of our calendar is one of confusion so let us start with some basics. From earliest times, anticipating the changes of the Seasons would have been vital for survival. The easiest way to do this would have been to count the number of lunar cycles. Accordingly, most early calendars were lunar based, with a lunar month being set as the length of time it took for the moon to complete its cycle of waxing & waning as it orbited the Earth. We now know that that lasts about 29.53 days so a good way of setting the months would have been to alternate between months of 29 days and months of 30 days. We now turn to the much-revered Sun. It had been seen that the Sun changed its position in the sky every day in a cycle, called a year, which embraced all four of the Seasons. It did not take long for astronomers to realise that a year corresponds with the rotation of the Earth about the Sun, or vice versa, and takes about 365.25 days. Accordingly, people had two main time systems: one based on lunar months and the other based on solar years. Unfortunately, they did not coincide. The nearest that a combination of synodic cycles would come to a single solar cycle would be 12 months but, with months alternating between 29 & 30 days, that would cover only 354 days. Different parts of the world took up different solutions to this problem. As we are focusing on Europe, we now look at what the early Romans did. They started with just 10 months in a year, of either 30 or 31 days length, plus an extra period over winter (apparently not thought of as a month) to make up the difference. They then developed a system in which years had 12 months of from 28 to 31 days length, totalling 355 days, with an occasional extra month added (between February and March) to make up the difference. It seems that political leaders were able to influence when the extra month was taken so, of course, the calendar soon moved out of synchronisation with the Sun. Julius Caesar put this right and set the year at 365 days every year except for every fourth (leap) year when it would be 366 days, thereby setting the average at 365.25 days. The new calendar, which was named the Julian Calendar after him, spread throughout most of Europe and, except for the changes reported below, remains with us today.

We have proved that we can live with a calendar of 12 unequal months but one issue complicated use of the Julian Calendar. There is an important word in the paragraph above: the “about” in “a year …. takes about 365.25 days”. A more precise figure is 365.242199. Over hundreds of years the difference between 365.25 and 365.242199 built up. Various religious festivals, in particular Easter, were traditionally set by reference to lunar months rather than calendar months and so this difference came to be noticed. By 1580, the accumulated difference meant that 10 extra days had been gained since the year of the crucifixion of Jesus Christ. This meant that Easter was being held 10 days later than some thought it should. Pope Gregory XIII threw his religious & political weight behind the need for a change to the calendar so that, when it was made (in 1582), the new calendar became known as the Gregorian Calendar. It is still known as that today. However, not every country was willing to follow Pope Gregory’s lead. In particular, it took time for non-Catholic countries to accept the practical benefits of adopting the Gregorian Calendar. The Kingdom of Great Britain and its dominions, then including most of the eastern states of North America, did not make the change until 1752 by which time the timing difference was 11 days. In England, Scotland, etc. the last day of the Julian Calendar was Wednesday 02.09.1752 and the first day of the Gregorian Calendar was Thursday 14.09.1752.

Moving from the Julian Calendar to the Gregorian Calendar involved:

(1) a change to the application of leap days, refining the rule of having an extra day in February when the year is divisible by 4 by not doing so when the year is also divisible by 400.

This still leaves a very small difference that will accumulate over the millenia but that was presumably accepted as ‘good enough’.

(2) an adjustment to the date to catch up with the difference that had built up over the hundreds of years since the Birth of Christ.

As mentioned above, in the countries which made the change in 1582 or soon thereafter,

the calendar had to jump forward by 10 days. For them, the last day of the Julian Calendar was Thursday 04.10.1582 which was followed by the first day of the Gregorian Calendar, Friday 15.10.1582.

(3) settlement of the first day of the year as 1st January, where this had not been done already.

More on this below.

Use of the Gregorian Calendar is now accepted throughout the world in almost every country. However, even though the leap year adjustment means that the day’s difference no longer accumulates in the way it used to, Easter and some other religious festivals are still set by reference to lunar months.

 

New Years Day

The beginning of Spring was long treated as the beginning of the Year but in 153BC the Roman Senate set 1st January as the date when new Consuls took power so that date came to be of practical importance throughout the Roman area of influence. Accordingly, when he set his new (Julian) calendar, Julius Caesar set 1st January as the first day of the year. However, that was not the end of the matter. Later on, particularly in the Eastern Roman Empire, 23rd September became treated as the first day of the year in memory of Augustus Caesar whose birthday it was, though it (the first day) was later moved to 1st September which remains an important date for many Orthodox Christians.

During the Middle Ages, the Christian Church’s opposition to the pagan New Year festivities grew. They imposed their own religious New Year, normally on 25th March as, having adopted 25th December as the birth date of Christ, they presumed that his conception had arisen exactly 9 months earlier. [25th March was named Annunciation Day though in some countries (including England) it was better known as Lady Day, probably an abbreviation of Our Lady’s Day.] However, this was not applied consistently throughout Europe (possibly because 25th March was noticeably close to the beginning of Spring and so could also be associated with pagan rituals). The (Western) Holy Roman Empire moved back to 1st January in 1544 and, over time, this was followed elsewhere. Scotland did not change until 1600 and England not until a remarkably-late 1752.

Consider the effects of this in England (similar examples could be shown for Scotland and other countries). Until 1751, the year lasted from 25th March until 24th March. So, for example, the day that came 4 days after 23.03.1700 was 27.03.1701. That might seem strange to us now but that was the convention that was used and, other than it being mid-month rather than at the end/beginning of a month, it is really not much different from our now holding that December is in one year but the next month (January) is in the next year. 1751 was a short year in England, lasting only from 25th March 1751 until 31st December 1751. There was, for example, no 15th March 1751. The 15th March that was 10 days before 25.03.1751 was shown as of 1750. The 15th March that came after 25.03.1751 was shown as of 1752. The change must have caused much confusion at the time. It can still cause confusion now when we look back to those times. Normally, we can simply ignore the problem and accept the dates that were used at the time so that, for example, 01.09.1551 under the Julian Calendar is treated as though it would have been 01.09.1551 under the Gregorian Calendar. However, the period 1st January to 24th March can cause difficulty. To borrow the example used by Wikipedia, the execution of King Charles I is normally shown nowadays as having taken place on 30th January 1649 but in contemporary documents the date was recorded as 30th January 1648. Two conventions are used to distinguish between the two systems:

1. the use of “Old Style” (OS) and “New Style” (NS) : Charles’s execution date could be shown as 30.01.1648 (OS) or 30.01.1649 (NS).

2. showing the date with a back-slash between the 2 possible years : Charles’s execution date could be shown as 30.01.1648/9.

Unfortunately, many databases (including Stirnet’s own Families Database) do not follow these conventions but, ignoring the first one, confuse the second one with the very-different convention on how to show uncertainty on dates. Under this latter convention, 30.01.1648-9 signifies that the date is 30th January but with uncertainty on whether the year is 1648 or 1649. Similarly, 1645-9 shows that the relevant year was in the range 1645 to 1649 inclusive.

Mea culpa : I do not always follow the back-slash convention (eg. 1648/9 being OS/NS) not least because, at an early stage in the development of the database, frustration with the whole issue of dates (disagreements between sources, etc.) led to the decision not to even attempt to be pedantic with dates but to view them as little more than guidelines, being accurate when it was reasonably easy to be so but not wanting to give the impression of reliability when such appeared not to be achievable. I have since modified that approach, and am now trying to comply with these conventions, but the damage done in the past will take years to repair whilst many of the sources I use are also not consistent with the way they show dates which often makes it difficult for me to be sure what they mean. Accordingly, the Families Database is not to be treated as consistent in its applications of the conventions so that, for example, 1648/9 and 1648-9 should be viewed as the same, both signifying that the exact year is uncertain, being either 1648 or 1649.

 

CONCLUSIONS AND SUGGESTIONS

1. IF YOU WANT TO WRITE A HISTORICAL NOVEL BASED ON ENGLAND IN THE REIGN OF GEORGE II (PARTICULARLY IN 1751), OR ON SCOTLAND IN THE REIGN OF JAMES VI (PARTICULARLY IN 1599), BE CAREFUL WITH THE DATES. ESPECIALLY FOR THOSE WHO HAD TO DO BUSINESS IN BOTH SCOTLAND & ENGLAND BETWEEN 1600 AND 1752, THE CHANGES MADE IN THOSE YEARS MUST HAVE CAUSED EVEN MORE CONFUSION TO THE PEOPLE OF THE TIME THAN GOING METRIC DID TO US NOT LONG AGO. IT IS STRANGE THAT THESE PROBLEMS RARELY GET A MENTION IN HISTORY BOOKS.

2. IF BEING PRECISE ABOUT A DATE IS IMPORTANT TO YOUR RESEARCH, DON’T FORGET THE ABOVE ISSUES OR ELSE YOU MIGHT FIND THAT YOU RECORD OR INVESTIGATE THE WRONG YEAR. CHECK DATES YOU PICK UP FROM HISTORY BOOKS AS WELL AS ONLINE & OTHER SOURCES. PRACTICE WHAT WE PREACH RATHER THAN WHAT WE HAVE DONE AND, WHERE POSSIBLE, FOLLOW THE CONVENTIONS (CF. 1648/9 VS. 1648-9) AND TRY TO BE CONSISTENT IN DOING SO. INCONSISTENCY NOT ONLY MAKES CONVENTIONS INEFFECTIVE BUT ALSO CAN GIVE A FALSE SENSE OF RELIABILITY.

 

 

Postscript 1

The change of date on which New Years began appears to have been accepted reasonably easily. At a time when most people worked in agriculture or in service, more attention would have been paid to days, weeks & months than to years whilst the conventions reported above would have helped to ease the problems in dealing with the years. Most people would not have cared all that much whether or not 1st February was denominated as of 1751 or 1752 as long as they knew it was still called 1st February. However, the change of calendar in 1752 is reported to have upset some people because 11 days were taken out of the calendar altogether and they were told that the date which they were going to call 3rd September was instead to be called 14th September. Some people thought that 11 days had been taken out of their lives!

[If anyone has a copy of a newpaper article or similar from 1751-2 which discusses this issue, I would be very grateful if they would send me a copy of the text of it so that I can share it with others through this page.]

Postscript 2

In the UK, 6th April still marks the beginning of the year for tax purposes. This came about because it was based on 25th March, the old first day, plus 11 days for the Julian-Gregorian calendar conversion plus 1 day for the dropped leap day in 1900.

 

Sources

Various web sites including Wikipedia, most found by googling ‘Julian Calendar’, ‘Gregorian Calendar’ and ‘New Years Day’.