The year 0 AD

Alex Livingston alex at
Mon Jan 6 07:57:19 UTC 1997

>"Markus G. Kuhn" > INTERNET:kuhn at
>>ISO 8601 provides no way to represent years before the year 0000, or
>>after the year 9999.  This makes it difficult to represent timestamps
>>in some historical applications.  To fix this, you might extend the
>>syntax for date-fullyear to:
>date-fullyear = ["-"] 4*DIGIT
>>where the years are numbered ..., -0002, -0001, 0000, 0001, 0002, ....
>>(Note that unlike the traditional Julian calendar, there is a year 0
>>in the modern Gregorian calendar.)
>WRONG!  Negative year numbers are not part of the Gregorian calendar.

Still not quite right: *non-positive* numbers are not part of the Gregorian calendar. Which presents me an opportunity too hard to resist to scream something that I have never yet explicitly read anywhere on this discussion group or anywhere else on the Internet. Year numbers are *ordinal* (as opposed to cardinal - correct me if I've got this the wrong way round), i.e., like the day of the month (and the month of the year when written numerically), they indicate the "howmanyeth" year it is (I can't think how better to put it). AD 1 was the *1st* year (of the era). 1997 is the 1997th year. It follows that the *1st decade* comprised the years 1 to 10 *inclusive* (the 1st, 2nd, 3rd, ..., and 10th years); the *1st century* comprised the years 1 to 100 *inclusive*; the *1st millennium* comprised the years 1 to 1000 *inclusive*. And 2000 will be the 2000th year and hence the last year of the 200th decade, the 20th century, and the 2nd millennium.

This is one virtually unchallengable way of explaining why the next century (and millennium) begins at the beginning of 2001 (and not 2000) - and pointing that out is the main reason I'm writing this (though I guess I'm preaching to the converted here). Unlike time of day, specifications of year, month, and day of the month (and dates in general) indicate a *period* of time, not an instant. It is 1997 for a whole year; it is 10 o'clock for no time at all. Dates are discrete, time of day is a continuum. Just as the days within a month (and the months within a year) are numbered starting at 1, so are the years of an era. Another illustration (in case it still hasn't sunk in): The second half of a 30-day month starts on the 16th; the second half of the year starts with the 7th month; and the second half of a century starts with the 51st year of that century (ignoring for the moment the fact that months and years are not equal in length etc., making the use of the word "half" here questionable).

>There are two ways to number years before 1 AD, astronomical and historical.
>In historical dating, the year 1 AD is preceded by 1 BC, as there was no
>concept of 0 in Western civilization when Exigius set up the Anno Domini year
>count in AD 532.

The BC era is anomalous only in that the years are counted from latest to earliest rather than earliest to latest.

Whether or not there was a "concept of 0" back then is entirely irrelevant. (I'm sure that in fact there already was something akin to that - it just might not have been formalized.) With what number do you label the first of something? Go on, say it! Which month of the year is this? On what date did it begin? Which century did Jesus Christ die in? With which year of the era did the era begin?
>In astronomical dating, there is no BC: everything is AD, the year 1 AD is
>preceded by the year 0 AD, which is preceded by -1 AD, preceded by -2, etc...
>Here's a way to remember it: Julius Caesar was assassinated on 15 March 44 BC,
>historical dating, and -43 March 15, astronomical dating.

Introducing a year 0 in order to make centuries of the era start when three digits change in the year number sets the cat squarely among the pigeons:

If the BC era is to be retained and its centuries are also to change on such occasions, a year 0 BC would also have to be introduced! Then we'd have the years passing thus: ... BC 1, BC 0, AD 0, AD 1, ....  A fine improvement! Unless the year AD 0 is called the "1st year" (rather than the "0th"), which I would have thought rather defeats the purpose, calling the years 0 to 99 the "1st century" would be inconsistent anyway: they should be referred to as century 0 (the 0th century?), and the years 0 to 9 as decade 0, and the years 1000 to 1999 as millennium 1 (the 1st millennium?) and the years 1900 to 1999 as century 19 (the 19th century?).

If astronomical dating were to become the norm, what sense does a term like "1st century" make? What years did it consist of? Years 0 to 99 or years 1 to 100 (or years -14999999999 to -14999999899)? What was the century before it? The "0th"? What years did it consist of? Years -100 to -1 or years -99 to 0?

Facts must be faced (though some would disagree with me there I'm sure): years, months and days are numbered like lines of poem, verses of a song, pages of a book, lots in a subdivision, diskettes in a box, and countless (pardon the pun) other things. That this means that the first 10 of anything includes the 10th, the first 100 the 100th, the first 1000 the 1000th, and the second 1000 and the twentieth 100 the 2000th simply cannot be got around. By all means celebrate the beginning of the 2000th year at the beginning of 2000; but let that be the beginning of a year-long party climaxing at the end when the 2000th year, the 200th decade, the 20th century and the 2nd millennium end and the 3rd millennium, the 21st century, the 201st decade and the 2001st year begin. (On the other hand these are just arbitrary numbers given very tenuous significance only because of the human race's extremely unfortunate choice of base 10 to count in - and Jesus Christ was almost certainly born in BC 7 or 6 anyway, so we're celebrating 2000 years since nothing in particular!)

Macintosh Support
Computing & IT (C&IT)
Australian Graduate School of Management (AGSM)
The University of New South Wales (UNSW)
Sydney  NSW  2052

E-mail   : alex at; cit at (C&IT)
Facsimile: +61 2 9313-7279
Telephone: +61 2 9931-9264

More information about the tz mailing list